So lets review the clues. First was the photomosaic of the smile itself.
Clue 2, which I dropped even before clue one, but didn't make explicit unit this morning, was the presence of alders. As a general rule, alders like to have their feet wet. They indicate plenty of water. Not every time, but by and large, if you have a healthy stand with lots of them together, as you do here, you can bet there's plenty of water- if not on the surface, near the surface.
Clue 3 was probably not too helpful until I made the link between the cinder cone above and the basalt flow here.
working in the area in 1987, and during summer visits still find them very refreshing. But in the context of this puzzle, I'm pretty well convinced this cinder cone is the source of the lava flow that was quarried for road metal at the Cheshire Cat outcrop, as Ron Schott suggested in a comment, and illustrated with another cinder cone/lava flow pair from central Oregon. I also emphasized the permeability of this pile- if air can flow into and through it so easily, so can water.
Clue 6 was to point out the the lava flow is very different from most of the rock in the area: it is fresh and unaltered, so it must be younger than the hydrothermal alteration episode in the area, which is tied to an intrusion of granite-like rocks at 18 Ma, and which are exposed at Yellowbottom falls. Furthermore, especially near the heart of the district, the rock is very tough and not very permeable. Joints and faults may allow some groundwater movement, but most of the rock in this area is pretty tight. So one might expect most small streams to go fairly or completely dry by summer's end- no springs or runoff- and indeed, small ones do. But as I pointed out in clue 7, a smallish- smaller than the drainage of Dry Gulch- basin the size of Boulder Creek is large enough that the creek always has some flow in it. So one would expect that a larger one in the same general area would too.
Clue 8 was an interpretive sketch of the geology- in vertical cross-section- of the geologic units in the vicinity if the CC quarry. You can't see the base of the flow, but you can see the hackly exterior portion that cooled quickly, and the columnar jointed, slower cooling interior, and you can definitely get the sense of a valley in cross section. 9, the clue that finally made the connection for Eskered, was merely to explicitly point that "alders" was a veiled clue, which I had dropped so casually no one had caught it.
There was one more that I *could* have used, one that I often did when I was doing regular trips with middle schoolers, which is the question, "What do you find in streams and rivers around here besides water?" the answer is cobbles, gravel and sand. I was hesitant to use that one for a couple of reasons: first, when I was doing these trips, we'd have just spent the whole day looking at that sort of material, in and around the Quartzville Creek area- it was a logical leap. An online audience, I felt, might be more confused than helped by that question. Second, when you have a captive audience of 10-15 kids, you can bounce their own ideas off the rest of the group, and the feedback is much quicker and free-flowing. I didn't think those sort of quick questions and answers would work very well in a blogging environment.
As I commented in my response to Eskered's underestimated answer of "seepage," there is a very clear layer of cobbles below the road across from the CC quarry, and water is coming out everywhere for a stretch of maybe 15 to 25 feet. It's not easy to clamber down to it; the hill is steep with unstable, leaf littered soil and rocks. But the alders present there as well have good sturdy strength, and provide stable handholds and stops. I don't really have a clue how to estimate the total amount of outflow without a heck of a lot of work, because it's dispersed over a wide area, and I'm not experienced enough to eyeball it in any meaningful way. However, I'm confident that my claim of "many gallons per second" is simultaneously vague enough and satisfactorily accurate to be useful... it might be more precise (or not) to say "on the order of a few tens of gallons per second," but I'm still aiming at a vague but satisfactory description.
When I went up there again with GeoHols and her husband in early August, I did climb down closer to try and get a few photos of the springs. But once again, it was the end of the day, I was tired, and just didn't feel like pushing myself hard enough to go those last few feet and get all the goddamn vegetation out of the way between me and the gushing water. I will say, though, if you're in the area during hot weather, this is a nice spot to know about- I'd trust this water to be pathogen-free, it's icy cold, and the flow rate is quite high, overall. I suspect that the base of the cinder cone acts as a reservoir of sorts, because the flow has been quite high even during the end of summer/beginning of fall- there's not as much seasonal variation as one might expect.
Dana and I have decided that when and if we get back up to this area, we're going to do the trip backwards. This spot and the quartz vein near Red Heifer pass deserve to be taken with a full reservoir of energy, rather than waiting for the end of the day, when I'm just too tired to deal with them.
Also, following up on Ron's comment, there is a very clear outcrop of more of this basalt on the opposite side of Canal Creek, but I didn't think to get a shot of that. Canal Creek has cut several feet through the base of the flow, back into the older altered rocks of the area. Two things I should look for on future visits are whether there are cobbles and gravel under that exposure as well- that would tell me whether the ancestral Dry Gulch continued past the current position of Canal Creek, and thus whether Canal Creek eroded through the flow in more or less the same place it existed previously, or closer to the cinder cone than before it erupted. It also dawned on me fairly recently- in the last ten years or so- that this flow likely dammed a fairly large lake upstream from here, and it would be interesting to look around and see if I can find stuff that looks like lake sediments. They'd be pretty obvious and distinctly different from the other rock materials in the area.
Oh, yes, chain of events- I'll add dates, because I kind of can, with limited accuracy, not because the dates are important to understanding the sequence.
- Deposition of the volcanic and volcaniclastic rocks of the so-called Sardine Formation ~35-25 Ma
- Intrusion of granodiorite, hydrothermal alteration of Sardine Formation ~18 Ma
- Up to ~2 Ma, erosion, development of ancestral Dry Gulch basin, deposition of cobbles and gravels in stream bed.
- Sometime after beginning of Pleistocene (Cinder cone deposits overlie glacial debris, though I've not seen that outcrop myself), eruption of cinder cone, subsequent basalt flow, burial of the ancestral lower reach of Dry Gulch.
- Recent: During periods of very high runoff, the subterranean drainage of the ancestral Dry Gulch can't cope with the flow, so modern Dry Gulch has been eroded into what appears to be the northern margin of the basalt flow, but only on rare occasions actually has flow visible to an observer. (My own feeling is that these must be extreme enough events that I wouldn't feel safe being there during them.)
So there you have it... it turns out that the grin without a cat has a heck of an engaging tale to tell!