Forty years ago just before Christmas I was driving from Austin, Texas, to Vail, Colorado, to go snow skiing. As I inched along the icy highway leading toward Raton, New Mexico, a fierce snowstorm swept down across the high plains and nearly blew my VW off the road. Ahead through the blinding snow and twilight of late afternoon was a small side road and a sign saying, "Folsom 8 miles". Incredible! Just eight miles to the right was the exciting archaeological site that opened the door to our search for how and when the First Americans began to colonize the New World.

As I slithered along the icy side road, my mind flashed back to the textbook and archaeology course I had just completed, and to the existing theories about our ancient ancestors we called "Folsom, Clovis, and pre-Clovis". Ahead, under the overcast sky and blowing snow lay a small group of abandoned stores, a run-down hotel, some weather-beaten frame houses, and a tiny grocery store with two pick ups parked in front that doubled as a post office and gas station. A block away and across the street loomed an old frame building with a wooden sidewalk in front and a battered sign hanging at an angle by a single nail that read: "Museum".

On the inside of the frosted-glass door was a small sign that said, "open". As I entered, I noticed that I was alone in a dimly lit, unheated room full of displays of pioneer farming days to World War II relics. Scattered on tables, hanging on walls, and in various dusty display cases were tools, broken toys, military uniforms, old dresses, a display of rusty guns, faded black and white photos, and a corner display of what a kitchen might have looked like during the mid 1800s. Open on a small table, next to an empty pickle jar with a faded sign saying "donations", lay an open guest register with curling pages; its last entry was more than a six-months old.

In a far corner of the next room I saw the skull of an extinct bison hanging on the wall with rusting shovels and other excavation tools below it in a roped off area. Next to it was a hand-labeled display laid out under the clouded glass of a display case. Neatly arranged in rows were stone artifacts and badly-faded photos of the 1920s excavation of the nearby Folsom site that revolutionized American archaeology.

In the third room were more stone artifacts, mostly from modern Indian groups and a large painted mural on the back wall. There, almost life like was a painting of a large, wounded mammoth surrounded by lightly-clad hunters hurling spears. A sign on the wall said "Clovis Man".

As I walked around on the creaking, wooden floors and shivered in the cold, I read the many hand-written labels in other display cases. It was an eerie place and I felt as if I had stepped back in time. Each breath left a small cloud in the cold air and the only sounds were the howling wind outside and the banging of a distant window shutter.

After closing the front door I just stood there on the wooden sidewalk staring down the street at what once was a bustling community, but had long ago been all but abandoned. I was overcome by wonder. Was I really here? Was this really the place that my professor had called, "the most important archaeological breakthrough in the history of North America"?

I didn't know it then, but I was hooked. In the years after my visit I would go on to finish my education and then be part of the search for early Americans at sites in Texas such as Bonfire Shelter, Lubbock Lake, and Gault, in the Northeast at Meadowcroft, in the Pacific Northwest at the Marmes Site, and in Peru at the site of Pikimachay. I worked at those and other sites not as an archaeologist, but as a palynologist, a person searching for microscopic clues of fossil pollen that might help us understand the paleoclimate, subsistence patterns, and ancient vegetations that affected the lives of early Americans and the animals they hunted.

During the past two decades pollen research has become common and expected at most archaeological excavations, but this has not always been true. Initially, archaeologists were slow to include pollen studies as part of their excavation plan and were slow to embrace the importance of the data fossil pollen could provide. Part of the hesitation was a lack of understanding of how pollen data might help interpret the past and a lack of understanding of how and why fossil pollen studies were valid avenues of scientific research.

A good place to begin our story of pollen is with a lecture presented in 1916 at a meeting of Scandinavian scientists. At that meeting, Lennart von Post, a Norwegian geologist shocked his colleagues by saying that the recovery of fossil pollen from buried sediments was the most precise method yet developed for interpreting past vegetational regimes and determining cycles of vegetational change. Scientists had seen fossil pollen in ancient deposits, but no one had found a way to interpret them effectively. As he lectured, von Post set forth the basic theory of pollen analysis and explained why pollen was the ideal tool for studying changes in past vegetation, and by inference, climate. First, he pointed out that many plants produce great quantities of pollen or spores that are dispersed by wind currents. Second, he noted that pollen and spores have very durable outer walls that can often remain preserved for thousands or even millions of years. Third, his research had indicated that the unique morphological features of each type of pollen and spore remains consistent within each species, yet each different species produces it own specific form. Fourth, as ecologists had already discovered, each pollen and spore-producing plant is restricted in its distribution by environmental conditions that include moisture, temperature, and soil type. As such, each species is most plentiful in areas that best meet the plant's optimal needs. And fifth, von Post determined that most wind-dispersed pollen and spores rarely travel very far before falling to the earth's surface within a small radius (within 50 km) from their dispersed source. Thus, by counting a sufficient number of fossil pollen and spores recovered from each stratum in a deposit, one could reconstruct the types and abundance of plants represented by those fossil grains.

Using the principles, he set forth in his lecture, von Post then detailed how he was able to use his pollen studies of bog deposits in central Sweden to reconstruct the sequence of vegetation changes for that region. He pointed out that his data detailed thousands of years of change beginning with the early vegetation of pioneering plants that grew in the region immediately after the continental glaciers receded, through various stages of forest succession, and ending with the present climax forests of spruce and pines. Subsequent research confirmed the validity of von Post's research.

The magnitude of pollen production by some plants staggers the imagination. Many plants rely upon the wind to carry spores or pollen to their intended destinations, yet wind pollination is an inefficient method. Thus, to insure fertilization plants must produce great volumes of pollen in hopes that at least a small fraction will find its intended destination. So great is the pollen production of conifer trees, that current Swedish scientists estimate the forests in the southern third of their country annually disperse over 75,000 tons of microscopic pollen into the atmosphere. Heavy pollen production is not limited to conifers. For example, in the United States plants such as marijuana (Cannabis) produce over 70,000 pollen grains per anther and a single branch on a male marijuana plant can produce more than 500 million pollen grains.

Around 100 million years ago during the Cretaceous Period, a large number of plants began to develop more efficient methods of pollen dispersal that relied on insects and small mammals instead of the wind. With this change came a vast reduction in the need to produce pollen grains. This reduction became so great that some of the modern, insect-pollinated plants such as clover need to produce no more than 200 pollen grains per anther to insure pollination. Others in this group, including maple trees, have found that around 1,000 pollen grains per anther are ample enough to ensure proper seed production. These types of plants now far outnumber the ancient wind-pollinated types and now form the major components of many plant communities.

Almost all of the pollen von Post found in his analysis of Swedish peat deposits was from wind-pollinated plants. Herein lies one of the limitations of pollen analysis. Pollen records are excellent capsules of information about which "wind pollinated" species once lived in a region, yet these same records tell us almost nothing about the insect pollinated plants that were also present. This imbalance of preserved pollen information is not as limiting in some environments as it is in others. For example, most of the vegetation in boreal forests is wind-pollinated, therefore the fossil pollen record captures a fairly good image of that ancient vegetation. However, in other regions, such as those dominated by tundra and deserts, most plants are insect-pollinated and therefore those deposits contain a very sketchy fossil pollen record.

Many Paleoindian sites in the New World do not contain fossil pollen because of their environmental location. Rather than try to explain why this phenomenon occurs, I think that subject is sufficiently complex to cover in a separate article. Instead, I would rather focus on the Paleoindian site of Bonfire Shelter and show how fossil pollen solved an important mystery.

Bonfire Shelter is a unique site in Southwest Texas located hundreds of miles south of the Great Plains. During the Late Pleistocene and Holocene that region of southwest Texas was outside the normal grazing range of most bison herds, yet Bonfire Shelter is famous for being a site where large herds of ancient bison were stampeded over a cliff to their death.

When excavated, Bonfire Shelter presented a puzzle for archaeologists. In the bottommost cultural deposits there were broken bones of Pleistocene megafauna and hints that they had been killed by Paleoindians. In two separate intervals above those deposits there are thick layers of fossil bison bones. Each of those two zones contains the remains of hundreds of bison that were driven to their deaths in multiple jumps. Archaeological evidence suggests that the bison jumps took place fairly quickly and that the bones in each of the two intervals accumulated during a period of no more than about 100 years.

The lower, thick deposits of bison bones are dated to the Folsom era and are associated with artifacts from that Paleoindian period around 10,000 years ago. The upper bison bone layer contains stone tools from fairly modern Indians and the deposits have been dated as being around 2,500 years old. What puzzled archaeologists was why Bonfire Shelter had been used as a bison jump site during only two short intervals? Also puzzling was why, for the 7,500 years in between those bone deposits the site has been completely abandoned?

Fossil pollen studies of the deposits in Bonfire Shelter and from sediments in other nearby archaeological sites revealed the answer. It seems that during the past 12,000 years the vegetation in southwest Texas was initially cooler and contained sufficient grass and brushy vegetation to support various species of Pleistocene megafauna. Later, the brushy vegetation was replaced by grasses that reached their maximum coverage only twice. In other words, local grazing conditions were ideally suited for large bison herds only twice, each for only a brief interval, during the last 10,000 years. During each maximum grass period bison jumps occurred at Bonfire Shelter. Strangely, at other nearby archaeological sites, which were occupied during both of these maximum grass intervals, bison bones are absent or exceedingly rare. These data suggest that even though bison were plentiful, they were not hunted by local Indians living in the Bonfire Shelter region. Instead, archaeologists now believe that only skilled, nomadic hunters who followed the bison herds south understood how to kill bison and twice they discovered that the cliff above Bonfire Shelter was an ideal location for bison jumps. At other sites near Bonfire Shelter it appears that local groups hunted smaller animals such as deer, rabbits, and a variety of small rodents throughout the last 10,000 years.

The pollen records at Bonfire Shelter are important for another reason. The Devil's Mouth Site is located on the banks of the Rio Grande River 60 miles southeast of Bonfire Shelter. When it was first excavated, radiocarbon dating was not possible for most of the upper strata. Fortunately, fossil pollen was preserved in those upper zones at the Devil's Mouth Site and the pollen types were similar to those found at Bonfire Shelter. By matching similarities in both pollen records, it was possible to assign estimated dates for some deposits at the Devil's Mouth Site. More than a decade later when renewed excavations at the Devil's Mouth Site uncovered charcoal-filled hearths, the resulting radiocarbon dates revealed those deposits were less than 100 years different from the estimated dates previously assigned by them by the pollen record. Fossil pollen cross-dating of archaeological sites does not always work this well, but when it does it is impressive.

Unfortunately, few of the earliest sites associated with the First Americans and Paleoindians have been thoroughly tested for fossil pollen. Of the few sites where fossil pollen studies have been conducted, some, such as Bonfire Shelter, have yielded stunning results. However, at most sites the pollen results have been inconclusive or controversial. In Part II of this story (later newsletter) I will examine the problems of pollen sampling, pollen preservation, and pollen interpretations as they apply to the myriad sites in the New World associated with the First Americans and other Paleoindians.

---

  This article first appeared in CAP Newsletter 26(1), 2003.