Apollo Expeditions to the Moon



The last crystallization age of some of the Apollo 16 rocks appeared to be about 3.9 billion years, and continued to indicate that this age is a major turning point in lunar history. This general age for the cooling of highland-like materials also was found to hold for the ejecta blanket of the Imbrium Basin at Fra Mauro, for the rocks of the Apennines, and later for some of the highland rocks at Taurus-Littrow. This limit suggested (1) a major thermal event associated with the formation of several large basins over a relatively short time, or (2) a major thermal event associated with the formation of the light-colored plains, or (3) the rapid cessation of the period of major cratering that continually reworked the highlands until most vestiges of original ages had disappeared and only the last local impact event was recorded. As we attempt to explain the absence of very old rocks on Earth, we should not forget these possibilities for resetting our own geologic clocks.

Photo of Apollo 15 CSM Endeavour The eyes of geochemical sensors peer through an opening in the sides of Apollo 15 CSM Endeavour. The instruments gave us broad-scale remote sensing of the lunar surface, allowing data from sampies collected on the surface to be correlated across major areas of the Moon. Included were precision cameras and spectrometers that sensed x-rays, gamma rays, infrared radiation, and the chemical ions in the ultrathin lunar atmosphere. Apollo 15 cameras supplied much imagery used to plan our Apollo 17 exploration.

A Photo of astronaut, Ron Evans
Adrift between the Earth and the Moon, Ron Evans retrieved the film canister of the mapping cameras on the day after Apollo 17 left lunar orbit. His space walk lasted an hour, and resulted in the successful retrieval of data from three experiments. Ron's oxygen was fed from the spacecraft through the umbilical hose, with an emergency supply on his back. I was in the open hatch to help in retrieval, which was necessary because the service module would be jettisoned before we reentered the Earth's atmosphere.

Apollo 16 continued the broad-scale geological, geochemical, and geophysical mapping of the Moon's crust from orbit begun by Apollo 15. This mapping greatly expanded our knowledge of geochemical provinces and geophysical variations, and has helped to lead to many of the generalizations it is now possible to make about the evolution of the lunar crust.

Apollo 17 carried Capt. Eugene A. Cernan, Capt. Ronald Evans, and me in December 1972 to the valley of Taurus-Littrow near the coast of the great frozen basaltic "sea" of Serenitatis. The unique visual character and beauty of this valley was, I hope, seen by most people on television as we saw it in person. The unique scientific character of this valley has helped to lessen our sadness that Apollo explorations ended with our visit. It would have been hard to find a better locality in which to synthesize and expand our ideas about the evolution of the Moon.

Left image Right image 2
Apollo 17 photographed itself in the frame at right from its panoramic camera, which shows the valley of Taurus-Littrow after the landing of Challenger. The landing point (see inset at left) is revealed by a bright spot, produced by the effects of the descent-engine exhaust. A reflection from and shadow of the LM are also visible in high-quality prints. The panoramic camera has an absolute resolution of about 1 meter in best prints.
Detailed analysis of this panoramic photograph indicates that the light-colored avalanche, and many of the craters on the valley floor, are probably the result of the impact of material ejected some 50 million years ago from the crater Tycho 1300 miles to the southwest. Such orbital data have been invaluable in expanding the context of our interpretations of samples and data that were returned from explorations on the surface.

At Taurus-Littrow we looked at and sampled the ancient lunar record ranging back from the extrusion of the oldest known mare basalts, through the formation of the fragmental rocks of the Serenitatis mountain ring, and thence back into fragments in these rocks that may reflect the very origins of the lunar crust. We also found and are now studying volcanic materials and debris-forming processes that range forward from the formation of the earliest mare basalt surface through 3.8 billion years of modification of that surface.

The pre-mare events in the Taurus-Littrow region that culminated in the formation of the Serenitatis Basin produced at least three major and distinctive units of complex fragmental rocks. The oldest of these rock units contains distinctive fragments of crystalline magnesium and iron-rich rocks that appear to be the remains of the crystallization of the melted shell. This conclusion is supported by a crushed rock of magnesium olivine with an apparent crystallization age of 4.6 billion years. The old fragmental rock unit containing these ancient fragments was intruded and locally altered by another unit which was partially molten at the time of intrusion about 3.9 billion years ago. Such intrusive fragmental rocks are probably the direct result of the massive impact event that formed the nearby Serenitatis Basin; however, an internal volcanic origin cannot yet be ruled out. The third fragmental rock unit seems to cap the tops of the mountains and it may be the ejecta from one of the several large old basins within range of the valley. This unit contains a wide variety of fragments of the lunar crust, including barium-rich granitic rock.

Photo of the crater Alphonsus
A landing site not visited, the crater Alphonsus is shown in this Apollo 15 mapping-camera picture. A point near the right edge of the crater floor in this southward-looking view was once the leading candidate for our landing site in Apollo 17. However, the need for greater geologic and geographic variety resulted in final selection of Taurus-Littrow instead. Our regret is that both could not have been explored.

The valley of Taurus-Littrow and other low areas nearby appear to be a fortuitous window that exposes some of the oldest, if not the oldest, mare basalt extrusives on the Moon. They are about 3.8 billion years old, and are 50 to 100 million years older than the basalts at Tranquility Base. They also contain titanium oxide in amounts up to 13 percent by weight.

Previous Next Index