Geological Survey Bulletin 1673 Selected Caves and Lava Tube Systems in and near Lava Beds National Monument, California

Craig Cave and Craig Temple

An oval collapse trench 125 ft long and 60 ft wide provides entrances for both Craig Temple and Craig Cave. Downstream from this breakdown most of Craig Cave is nearly drained of lava; the last trickle of pahoehoe lava to occupy the tube filled only the rounded central part of the tube's floor and seldom reached the steep walls except on the outside of bends. At its downstream end the lava tube is plugged by a roof collapse. In contrast, Craig Temple, the part of the tube upstream from the breakdown, is two-thirds filled by two tongues of lava. A late tongue of the basalt of Mammoth Crater that formed the tube advanced downstream. A tongue of the much younger basalt of Valentine Cave entered from the surface and flowed into the upstream end of the collapse trench (fig. 56); it then cascaded over the upstream slope of the collapse pile that lies at the entrance to Craig Temple.

Figure 56. Frozen tongue of much later basalt of Valentine Cave (see fig. 4 and map 8, pl. 3) flowed over edge of collapse trench that allows access to Craig Cave (left edge of photograph). Continued flow of this younger basalt would have filled Craig Cave (see fig. 4 and map 16, pl. 5) and buried trench. Craig Temple, upstream, was partly filled by the younger flow.

Craig Cave was named by J.D. Howard in honor of the Craig brothers, who first guided him to the cave. Some confusion exists in records at Lava Beds National Monument regarding use of the name "Craig Temple." It is reasonably certain that Howard applied this name only to the upstream continuation of the cave above the collapse trench, which he explored during a second visit. One can imagine a resemblance to a temple in the upper extension of the cave where two steeply sloping and rough-surfaced flow fronts produced a room 40 ft wide and 100 ft long, with an impressive domed roof rising above it. The extremely rough surface of huge tilted and upturned pahoehoe slabs on the flow that advanced upstream from the breakdown may be likened to pews in a cathedral. The downstream flow advanced in flow units and produced a series of altar-like steps. On map 16 (pl. 5), the name "Craig Temple" is used only for this room in upstream Craig Cave tube. However, at some time in the past, the words "Craig Temple" were painted on the rock above the entrance to Craig Cave. To add to the confusion, the pamphlet "Origin of Georaphical, Geological, and Historical Features in Lava Beds National Monument and Adjacent Lands" (available at the Visitor Center) states that Craig Cave "was originally called Craig Temple."

Craig Cave

An impressive cavern more than 50 ft wide yawns beneath a broad arch in the cave roof (fig. 57) at the downstream end of the collapse trench, which provides access to the Craig lava tube (map 16, pl. 5). In the roof of the cave, and in the north wall of the trench, are several well-defined flow units of basalt lava. Most are less than 4 ft thick and are similar to the flow units from the collapse trench at the downstream entrance to Post Office Cave. The Craig tube is probably a downstream continuation of the tube system that forms Post Office Cave. Collapse rubble half fills Craig Cave directly beneath the entrance overhang and slopes steeply downstream to the cave floor. At the base of this pile the ceiling height is 20-25 ft. From this entrance the lava tube extends southeast on a gentle gradient for 156 ft and then rounds a broad curve until, 145 ft farther, the tube turns due south. A blanket of collapse blocks from the ceiling masks most of the floor, but three large patches of pahoehoe are free of rubble. They are in the central part of the tube, and the floor rounds upward into the walls on either side of them.

Figure 57. Entrance to large tube containing Craig Cave (see fig. 4 and map 16, pl. 5) is roofed by numerous thin layers of lava.

The left (northeast) wall over this 240-ft stretch of tube has lost most of its dripstone due to collapse, but on the opposite (southwest) wall several layers of dripstone plaster are peeling off near the base. Some large areas of dripstone remain attached to the higher parts of the wall. Very few areas of lavacicles remain on the high ceiling (from 20 to 25 ft) because most have peeled loose and fallen to the floor.

For the next 450 ft downstream, the tube's course is a large smoothly rounded curve open to the northeast; then it changes direction to almost due east and maintains this easterly course for the remaining 580 ft of the cave.

In the area of the broad curve the patches of pahoehoe that appear from beneath the rubble are concentrated along the south (outside) wall of the bend, and the tube is deeper here than on the north wall, as in the deepest part of a channel (thalweg) of a meandering river. Moreover, the south wall is relatively free of collapse rubble, except for thick slabs of dripstone plaster that have peeled away from the walls; the ragged upturned edges of these slabs form thin, irregular bench-like projections along the base of the south wall. Patches of excellent clean pahoehoe in the floor exhibit ropes that reveal the curving direction of the last lava flow that swept around this large meander-like bend.

A similar but relatively straight stretch continues downstream on an easterly course. The tube widens to an average of 60 ft over much of this stretch. Patches of dripstone plaster are peeling from the base of the walls on either side. At approximately the middle of this stretch very large collapse blocks—as much as 20 ft long—are scattered over a large collapse pile. These must have come down in a massive rockfall instead of by the slow upward unraveling of the roof when vertical cracks and columnar joints gradually spread open and divide the flow units of the roof into small blocks. Lava stalagmites (small piles of lava accumulated by drippage from the ceiling) embellish the pahoehoe floor of the cave at several places.

At the downstream end of the accessible part of the tube a huge collapse pile rises in a steep talus slope 50 ft above the floor. The broad top of this collapse pile lies beneath a collapse cupola about 20 ft above the original roof of the lava tube. The top of this cupola intersects an interbed between two lava flows that consists of red tuffaceous material, chiefly silt and sand, which is 3-4 ft thick. That abundant tree rootlets extend downward through cracks into this interbed shows the surface is not far above.

The broad top of the collapse pile beneath this cupola reverses slope and descends to the east on a gentle incline. Within 20-30 ft, however, the collapse blocks crowd against the steeper slope of the roof of the cupola above and seal off further access. Near the center of the slope one can work down another 20-30 ft along a small crawlway between fallen blocks just under the roof of the cupola. As J.D. Howard wrote after his examination of the cave 50 years earlier, "the passageway is caved in at the east, but it possibly could be opened up."

Entrance Collapse Trench

In contrast to the huge arched cavern, which opens Craig Cave to the east, the entrance into Craig Temple (map 16, pl. 5) at the upstream end of the breakdown is nearly hidden.

The most interesting features of the collapse trench between the two entrances are three small lobes of younger lava of the basalt of Valentine Cave that spilled into the trench over its south wall. Each lobe must have been moving slowly, almost devoid of energy. A lava cataract that formed where the eastern lobe spilled over the near-vertical wall of the trench is perfectly preserved, still adhering to the breakdown wall (fig. 56). Only a small amount of lava made it to the floor of the trench, where it spread in arched roof above is more than 50 ft wide a tiny delta over nearby collapse blocks. The second lobe, which rounded the west side of the obstruction on the rim, was even more sluggish. It congealed in place as it began the drop over the rim.

A third lobe of molten lava spilled into the northwest end of the entrance trench. This lobe had more energy and supplied a larger volume of lava. The lava eventually congealed after half filling the large Craig tube 80 ft upstream. The lava stream forms half of the floor of the large room called Craig Temple. If the eruption feeding the basalt of Valentine Cave had continued slightly longer, both Craig Cave and Craig Temple would have been filled and completely buried.

Craig Temple

The part of the Craig tube that lies upstream from the entrance collapse is 240 ft long, but only 125 ft of its downstream segment forms the high-ceilinged room that Howard called "Craig Temple" (map 16, pl. 5). The remaining 115 ft is a broad low-ceilinged area, parts of which can be reached only by crawling.

Craig Temple was formed by the underground merger of two flows of lava. The younger one, the basalt of Valentine Cave, advanced upstream in the Craig tube over the apron of collapse rubble. Its surface is very rough and uneven and reveals how the flow subdivided into trickles and tongues among the collapsed blocks as it cascaded down the steep apron of rubble onto the floor of the Craig tube. The other flow was a late tongue of basalt of Mammoth Crater, the same basalt that built the Craig tube. It spread downstream from within upper Craig tube. This compound flow is the lava fill that blocks the tube 240 ft upstream from the entrance and ends downstream with a sharp junction against the younger upstream-flowing lobe of basalt of Valentine Cave. This junction is 160 ft downstream from where the tube is blocked by lava rising to its roof and is 80 ft upstream from the west edge of the entrance. This junction is the lowest area in Craig Temple; the arched roof above is more than 50 ft wide and 10 to 18 ft high.

The late Mammoth Crater flow within Craig Temple is not a single flow that filled the 25-ft-high Craig tube at one time. Instead it is composed of dozens of small flow units, most only 1-5 ft thick. They moved slowly and piled on top of one another as the flow advanced. The congealed flow fronts of some of the thicker flow units form a series of lobe-like steps, 3-5 ft high, that form the dais-like structure of the temple. Each successive bench near the upstream end of the cave becomes thinner and more discontinuous, and the stack of benches finally reaches the roof, blocking further upstream access. The lava seems to have been fed very slowly into the tube and to have lost its energy by the time it arrived at this section.