Letters From Grand Canyon—
Piracy and Capture Carve the Grand Canyon: Part A

In the previous Letter, “Reversal” (bqr 14:3), we saw how the ancient stream pattern on the southern Colorado Plateau may have changed—by means that we don’t understand very well—from flowing northward into the sump lakes in Utah to something resembling the present southerly and southwesterly course of the Colorado River. Now it is time to focus on the area—northern Arizona—and the subject—the age and formation of the Grand Canyon—that hold so much interest for us. This is a complicated story with many theories and counter-theories, so we will have to divide this Letter into three parts, to be published sequentially.
Early Views: A Simple Scheme
Early geologists like Powell, and Clarence Dutton who wrote the wonderful Tertiary History of the Grand Canyon Region, were greatly impressed by the erosion they saw everywhere on the Colorado Plateau. So much erosion, they reasoned, must have taken long time and must have started early in the Tertiary Period, maybe some sixty million years ago. Since the erosion clearly is caused by the Colorado River and its tributaries, the river and its Grand Canyon must be that old also. What’s more, uplift of the Colorado Plateau, into which the Grand Canyon is cut, must be equally old. The reason is simple: You can only cut a deep canyon if the land surface is high above sea level, because rivers cannot cut below the level of the sea.
Conflict
These views held sway for many years, but trouble started brewing in the 1930s and ’40s, when geologists working in the Basin and Range country west of the Colorado Plateau pointed out that basins in that region are filled with material deposited locally in closed depressions. This material contains no evidence for a major through-flowing river such as the Colorado. Along the course of today’s lower Colorado River, some of these deposits of interior drainage are as young as six million years or so, but in much of the Great Basin they are being laid down even today. Particularly troubling are young beds indicative of interior drainage that are laid down across what is now the course of the Colorado River at the mouth of the Grand Canyon: no Colorado River could have flowed through the mouth of the Grand Canyon as recently as six million years ago.
The grand simplicity of the early views was now disrupted by a grand dilemma: we knew that the Colorado Plateau contains evidence of a south-flowing river system that is tens of millions of years old, but we had also just learned that the Basin and Range country, downstream along the same river system, contains evidence that the river is at most a few million years old. This contradiction was highly disturbing, the more so because most people thought—and many still do today—that the course of a river is more or less immutable once established. This implies that all parts of a river are basically of the same age, and that what is true of a part of the river must be true of the whole. Such notions are in stark contrast to the view (to which I subscribe) that river systems can, and in fact are likely to, change with time, evolving into new configurations by interconnecting in new ways, all brought about by some external stimulus such as uplift or warping of the land. Implicit in this is the possibility that different parts of a river can have different histories and be of different ages.
Conflict Resolution: Attempts
The last several statements should wake up the astute reader with a start. Wouldn’t this be a way of getting around the grand dilemma? Maybe the upper part of the Colorado River, on the Plateau, is older than the lower part in the Basin and Range province?
“Too novel”, said those who supported the notion of “One river, indivisible and old”. To avoid the unpleasantness at the mouth of the Canyon, they tried several tacks. The colorful geologist Charlie Hunt not only held fast to the notion of an ancient Colorado on the Plateau, but even gave this river its present course through the Grand Canyon. The river would then have escaped the immovable object at the Canyon’s mouth by means of a remarkable pirouette: the notion was that the Colorado exited the Grand Canyon southward by way of Peach Spring Canyon, an ancient valley a good part of which is now buried by younger deposits. Problem solved, thought Hunt. Unfortunately, it wasn’t, because my good friend Dick Young came to show that even the oldest deposits in Peach Springs Canyon point to streams flowing north into the Canyon, and not south, away from it as Charlie would have it. Besides, avoiding the mouth of the Grand Canyon does not solve the problem, which is a pervasive one: as we now know, any possible continuation of the hypothetical river downstream from Peach Springs Canyon towards the sea is just as plugged up by deposits of interior drainage as the area near the mouth of the Canyon.
Another notion that has been widely circulated is that the Grand Canyon and the river through it are in fact very old, but the river ceased to function temporarily at the time of the interior deposits in the basin and range country because it either ceased to flow or became so overloaded that it could no longer carry material through the Canyon into the country downstream from it. The result was that the Canyon became filled with debris, much of which consisted of “rim gravels”, deposits carried by the ancient northward drainage system and best exposed along the Mogollon Rim. This would make the canyon as old as the rim gravels, maybe even older.
There are many problems with this concept. One is that the Grand Canyon, all steep rugged walls and short stubby tributaries, has the characteristics of a young landscape. In the terminology of geomorphologists—people who study landscape—he canyon is “immature” and “youthful”, not thirty, forty, fifty million years old. Then, the rim gravels are much older than the interior-drainage deposits, so can hardly be used to explain away the great dilemma. A third problem is that the gravel terraces at Lees Ferry, held up as being part of the ancient fill of the Grand Canyon derived from the rim gravels, in reality are no more than a few hundred thousand years old, not tens of millions, and contain much material derived from the San Juan Mountains country to the north, rather than material derived from the south. But the biggest problem has to do with how rivers work: depositing hundreds or thousands of feet of fill into a large previously-carved canyon requires very unusual, and probably unrealistic changes in circumstances. The Colorado River has an enormous drainage basin that contains many mountain ranges, so is unlikely to run dry, as proposed. In any case, the large basin would ensure a high probability of floods, which are extremely efficient at carrying debris, and in fact do nearly all the work even in “normal” rivers, those with permanent flow. Desert washes seldom carry water, yet the occasional floods they experience are entirely adequate to transport whatever debris is dumped into them.
Conflict Resolution: A New Concept
The geologist Eddie McKee studied the Canyon so long and so well that many of us consider him a sort of patron saint of that remarkable place. In the early ’60s, he decided it was time to do something about the great dilemma. He knew the Canyon as well as anybody, and this knowledge enabled him to identify two areas where important information was likely to be found. One was the Hualapai Plateau, including Peach Springs Canyon, where it should be possible to test Hunt’s fluvial pirouette. The other was the Pierce Ferry area, just west of the mouth of the Grand Canyon, where the infamous interior-basin deposits were alleged to lie.

He then did three things to achieve his aim: he persuaded the Museum of Northern Arizona to establish small grants for the study of these areas; then, he found two enthusiastic and foolish graduate students (Dick Young for the Hualapai Plateau, and me for Pierce Ferry) to do the studying; finally, he set up a symposium for 1964 when Dick and I would present our results and the handful of people who knew anything about the Grand Canyon would get together to try to come up with some coherent—maybe even sensible—story on How It All Happened.
At this point, a little reminiscing seems appropriate. In the early ’60s, the West was still the West, quite innocent of the current hordes of Californians, Texans and people from everywhere else. There was little in the way of retirement colonies or trophy-home settlements. St. George was a tiny Mormon town; Mesquite consisted of a couple of barns and maybe a farmhouse; Las Vegas occupied a few blocks; the north boundary of Phoenix was not far from Camelback; Flagstaff did not reach the Museum of Northern Arizona at one end, while the other Museum (the Club) was out in the boonies at the other. US 66 was the only east-west highway because no Interstates existed yet; Highway 164 to the Four Corners was a very long dirt road; you drove to Phoenix by going down Oak Creek Canyon, passing through a Sedona limited to Uptown and basically devoid of tourists and vortices alike. Verde Valley was almost empty. Glen Canyon Dam was being built and the concrete was coming by truck from Clarkdale through the Oak Creek road. Just a little later, Lake Powell started being filled and upper Lake Mead shriveled to a collection of puddles. No private individual had a four wheel drive vehicle; the plague of suv’s had not yet descended upon the land, and atv’s and dirt bikes happily had not yet been invented.
When Dick and I set forth on our respective missions, we went into a silent, empty, and untracked land. We got around mostly by shank’s mare, and considered ourselves fortunate to have a few fifteen minute topographic maps. Mostly, we had no maps at all. There were no other geologists around to speak of—this stuff was just too remote. Nevertheless, in due course we did manage to get our work done, the symposium was convened, the arguing concluded, and the results published by the Museum of Northern Arizona.
The symposium proceeded largely by elimination: the river could not have done this, gone there. This is a sensible approach because you have a much better chance of proving that something isn't than that it is. So, yes, an ancient river did make it as far as the Kaibab Plateau, but no, it did not leave the Grand Canyon near its western end. In fact, we felt, the river did not even cross the Kaibab Plateau, which seemed a formidable barrier. So, what did the river do, assuming it did not go underground or just vanish into thin air? Driven to some extent by a notable lack of alternatives, we proposed that the river followed the course of what is now the Little Colorado river but flowed in the opposite direction, that is, southeast. Eventually, the river joined the Rio Grande and emptied into the Gulf of Mexico.
To those outraged by such a notion, let me say that reversals in the flow direction of rivers are not that uncommon in geology. In most cases, the mechanism causing the reversal goes by the swashbuckling name of “piracy and capture”, whereby some vigorous stream extends itself through headward erosion far enough to tap some less-vigorous stream in mid course, suddenly stealing and diverting the unfortunate victim's water. Rivers and washes extend themselves this way when they have a steeper gradient, so more erosive power, than their neighbors. This—piracy and capture—was the big conceptual novelty introduced at the symposium, a novelty that suddenly had a chance of solving the grand dilemma by making it possible for different parts of the river to have different histories and ages. And this is just what we proposed.
The old, sluggish ancestral Colorado river had been flowing peacefully south then southeast into the Gulf of Mexico for perhaps tens of millions of years when strange events happened southwest of the Colorado Plateau: here, the restless movements of the great plates into which the earth's crust is broken produced a linear depression, the Gulf of California, which opened five to six million years ago where no gulf existed before. The narrow northern end of this gulf extended up to north of Bullhead City, well into what is now the lower Colorado River corridor and less than a hundred miles from the edge of the Colorado Plateau. And now we had the makings of great change. Having the Colorado Plateau, standing 5,000 feet or more above sea level, so close to the sea means that any stream developing into the western edge of the Plateau and draining into the Gulf would have a very steep gradient indeed, at least fifty feet per mile. The course of the present Colorado River in the lake Mead area and western Grand Canyon would have been especially favored: in the Pierce Ferry area, the river developed in the low spot of the pre-river basin; on the Hualapai Plateau, it followed the valley at the foot of the Upper Grand Wash Cliffs, enriching itself with waters draining northward from the Hualapai Plateau; farther upstream, the Hurricane fault provided a belt of shattered rock that was easy to erode. With such advantages, and the steep gradient that was its birthright, the new river extended itself vigorously into the western Colorado Plateau, creating in the process the beginnings of the western Grand Canyon. In due course, the river reached and breached the Kaibab Plateau. This done, the new river was positioned to tap the old and sluggish ancestral Colorado, capturing and diverting its waters. Invigorated by increased flow and still made powerful by a steep gradient, the now-integrated river cut down like a buzz saw, carving out the Grand Canyon in just a few million years.
Conflict Part II
This was a reasonable proposition that did not violate facts known at the time. But it had an Achilles’ heel, which lurked in the country of the low divide separating the drainage basin of the Little Colorado River from that of the Rio Grande somewhere near the present Interstate 40. If the ancestral Colorado River indeed went where we proposed it did before capture, it should have passed through this area. But experts in the geology of the region were quick to point out that no known evidence documents the former passage through here of a river like the ancestral Colorado. This by itself was not necessarily a fatal flaw, because river deposits that could once have been there may later have been removed by erosion. But this problem became serious when combined with an additional one, which is that deposits and ancient surfaces buried by the Bidahochi Formation (in the Hopi Buttes country) seem more consistent with streams flowing generally northwest, as does the little Colorado River, than with the easterly flow direction required by our hypothesis. Since the Bidahochi is Pliocene and started being deposited perhaps six million years ago, the erosion surface on which it deposited must be older, which places it squarely in the time when the ancestral Colorado should have been flowing eastward through this region. Regrettably, the hypothesis advanced by McKee and colleagues did not stand the test of known evidence; it was necessary to abandon it, at least in part.
Ivo Lucchitta
This is the seventh in a series of “Letters from Grand Canyon by Ivo Lucchitta that will appear in future issues of the bqr. This particular “Letter” will be divided into three parts.