A New History of Life (30 page)

The Carboniferous oxygen peak would have had consequences in addition to gigantism. Oxygen is combustible, and the more there is the bigger the fire; it facilitates fuel ignition, and the fuel in question was the huge and global forest of the coal age.

The Carboniferous period may have witnessed the largest forest fires ever to occur on Earth (at least until the dinosaur-killing and forest-igniting Chicxulub asteroid of 65 million years ago, that is). Like so much dealing with the change of oxygen over time, studies on the possibility of megafires provoked by high atmospheric oxygen have been controversial, but are becoming much less so as more and more evidence accumulates. Indeed, the forest fire controversy has been a major criticism of the entire theory that oxygen values have been different in the past (including higher). It was suggested that ancient forests would not have been able to survive the catastrophic fires, and since we have a long fossil record of the forest, the catastrophic fires did not take place.

Conditions of elevated oxygen at least theoretically should generate more rapid rates of flame spread, as well as lead to higher-intensity fires, and indeed large deposits of fossil charcoal in sedimentary rocks of Mississippian and Pennsylvanian age in North America
⁵
are evidence
that there were forest fires back then: forest fires that were larger, more frequent, and more intense than those of today, although direct comparison suffers from the very different biological makeup of forests then and now.

If there were more and more intense forest fires, we would expect to see morphological adaptations to fire resistance over time. Plants evolved a well-known series of adaptations collectively known as fire-resistance traits, which include thicker bark, deeply embedded vascular tissue (Cambria), and sheathes of fibrous roots surrounding the stem.

One can also question why such high oxygen did not cause all Carboniferous forests to burn to the ground. While fires then do seem to have been more frequent, the presence of fire-resistant plants and the high moisture content both in the plants themselves and in the swampy terrain of large portions of the Earth’s surface in the numerous coal swamps limited damage. Also important is the temperature of the “match” that started the forest fires. In recent studies
⁶
looking at oxygen levels and whether wood would burn, the investigators reported that plants will not burn at oxygen levels below about 11–12 percent. Yet they tried to start these fires with a lit match, rather than the far higher temperature than any lightning strike brings.

Like animals, plants need oxygen for life. Oxygen is taken up within the cells during photorespiration. But the levels are far lower than those needed by animals for the most part. A second difference is that various parts of terrestrial plants, for instance, have different oxygen needs. Most plants live in two very different media—part in air, part in solid (the roots in soil). The very different environments of underground roots, surrounded by water, solids, and gas, required very different evolutionary needs. Leaves sit in air. They worry about losing water and getting enough light, not drowning in too much water (if they could worry, that is). But mostly the roots have a requirement that the leaves do not—the right level of oxygen. It is the root system
that is most susceptible to damage or cell death from low oxygen, and this is all too familiar to those gardeners or house plant owners who water their plants too much. Roots live in the underground environment where low-oxygen conditions can occur, even at times of well-oxygenated air, especially if there is too much water in the soil. Roots can be smothered by groundwater with low oxygen values, for instance.

What about plants and high oxygen levels? Here there is far less data, but what is known suggests that elevated levels of O
₂
are deleterious to plants. Higher levels of oxygen in air lead to increased rates of photorespiration, but a more serious consequence is that in higher oxygen levels there are more toxic chemicals called “OH radicals” that are dangerous to living cells. To further test these possibilities, David Beerling, a former student of Yale’s Bob Berner, grew various plants in higher than current oxygen within closed tanks.
⁷
When oxygen levels were raised to 35 percent (thought to have been the highest levels of all time, occurring in the late Carboniferous or early Permian), the net primary productivity (a measure of plant growth) dropped by a fifth. It may be that the higher oxygen of the Carboniferous through early Permian caused a reduction in plant life to some degree, although this is not observable in the fossil record by any dramatic change or mass extinction during this interval.

The conquest of the land by the chordates, our lineage, required many major adaptations. Most pressing was a way of reproducing that allowed development of the embryo in an egg out of water. The amphibians of the Pennsylvanian and Permian presumably still laid eggs in water, and thus could not exploit the resources of land regions that were without lakes or rivers. The evolution of what is termed the amniotic egg solved this. Presumably, it was this egg that ensured the existence of a stock of vertebrates now known as reptiles. The evolution of the amniotic egg differentiates the reptiles, birds, and mammals from their ancestral group, the amphibians.

The fossil record suggests that the amniotes are monophyletic: that is, they have but one common ancestor, rather than this condition arising more than once. That ancestor, an amphibian, lived some time in the Mississippian, and thus this crucial transition took place as oxygen levels were rising. The first amniotic eggs were probably produced at oxygen levels equal or even higher than that of today.

Reptiles are also considered to be monophyletic, a single species stock that diverged from amphibian ancestors perhaps some time in the Mississippian period of more than 320 million years ago. As we have seen, this was a time of rising oxygen, and a time as well of a major diversification of land- and water-dwelling amphibians. But while genetic evidence of this divergence can be dated back to as long ago as 340 million years, fossils that are ascribed to the first reptiles (instead of terrestrial amphibians) have been recovered from several localities globally. Fossils of small reptiles named
Hylonomus
and
Paleothyris
have been found interred in fossilized tree stumps of early Pennsylvanian age, and it may be that the fossil record of this later appearance is more valid than the assumption of a Mississippian evolution of the group. In either case, these first reptiles were very small, usually only about four to six inches long.

The skulls of these first reptiles had no tympanum (eardrum) and thus they could not hear well, or at all, and unlike the labyrinthodont amphibians, they lacked the large pair of fangs that was found in most of the larger carnivorous amphibians. Compared to these huge amphibians, the first true reptiles had a postcranial skeleton adapted to provide better and surely faster locomotion. They had very long tails relative to their body.

That these forms laid the first amniotic eggs is still speculation. There are no fossil eggs in the stratigraphic record until the lower Permian, and this single find remains controversial. But the pathway to the amniotic condition probably passed through an amphibian-like egg (without a membrane that would reduce desiccation), but lay in a moist place on land. It would have been the evolution of a series of membranes surrounding the embryo (the chorion and amnion) covered by either a leathery or calcareous but porous egg that was
required for fully terrestrial reproduction. One possibility seemingly never mentioned is that these first tetrapods evolved live birth, so that the embryos were not born until substantial development within the female had taken place.

Eggs capable of successfully producing viable offspring eventually were produced on land, and it was to these new amniotic eggs that the level of oxygen and heat must have played a part. There is a huge trade-off in reproduction for any land animal using an egg-laying strategy. Moisture must be conserved, so the openings of the egg must be few and small. But reducing permeability of the egg to water moving from inside to outside also reduces the movement of oxygen into the egg by diffusion.
⁸

Without oxygen the egg cannot develop. It may be no accident that the evolution of the first amniotes occurred during a time of high oxygen. It seems inescapable that this reproductive strategy was and remains in those animals living at varying altitude, affected by atmospheric oxygen content, with higher oxygen contents producing more rapid embryonic development. High oxygen may have allowed live birth. Some biologists have suggested that live birth could not take place because, at least in mammals, the placenta delivers lower levels of oxygen than is present even in arterial blood in the same mother. But this generalization is for mammals only, where so much development takes place within an environment that can be regulated for its oxyen levels, temperature, and amount of liquid. Reptiles have a very different reproductive anatomy. It may be that low oxygen even favors live birth. Evidence to support this comes from three lines of evidence. First, it is well known that birds (egg layers) living in high-altitude habitats routinely feed at higher altitude than the maximum altitude where they can reproduce.

The maximum altitudes of birds’ nests of many mountainous species repeatedly show this pattern. The highest nests are at eighteen thousand feet, and higher than this the embryos will not successfully develop.
⁹
While at least three factors may be involved in this limit (lowered oxygen content with altitude, desiccation because of air dryness at altitude, and relatively low temperatures), it may be oxygen content that is most important.

Second, recent experiments by John VandenBrooks from Yale University have shown that alligator eggs taken from natural clutches and then raised in artificially higher oxygen levels showed dramatically faster than normal development rates. The embryos grew some 25 percent faster than the controls held at normal atmosphere oxygen levels. Increased oxygen clearly influences growth rates, at least in American alligators. Finally, Ray Huey of the University of Washington maintains that a higher proportion of reptiles at high altitude use live birth than do those at lower altitude.

As four-legged vertebrates emerged from their piscine ancestors, many new anatomical challenges had to be overcome. No longer was there water to support the animal’s body; in air, both support and locomotion had to be accomplished by the four legs. An entirely new shoulder and pelvic girdle design had to evolve, along with the muscles necessary to allow locomotion. Equally daunting was the problem of acquiring sufficient oxygen to allow sustained exercise. Early tetrapods apparently used the same set of muscles for motion and taking a breath, and they could not do both at the same time. Fish seem to have no problem with sustained exercise or with respiring during activity, suggesting that oxygen is not a limiting factor in daily activity. For land tetrapods, this is not the case. The body plan of the earliest land tetrapods provided for a sprawling posture with legs splayed out to the sides of the body trunk. In walking or running with such a body plan, the trunk is twisted first to one side and then to the other in a sinuous fashion. As the left leg moves forward, the right side of the chest and the lungs within are compressed. This is reversed with the next step.

The distortion of the chest during this kind of locomotion makes “normal” breathing impossible; each breath must be taken between steps. But this process makes it impossible for the animal to take a breath when running. Thus, modern amphibians and reptiles cannot run and breathe at the same time, and it is a good bet that their Paleozoic ancestors were similarly impaired. Because of this there are no reptilian sprinters. This is why reptiles and amphibians are ambush predators. They do not run their prey down. The best of the modern
reptiles in terms of running is the Komodo dragon, which will sprint for no more than thirty feet while attacking prey. This is called Carrier’s constraint, after its discoverer, physiologist David Carrier.

The dilemma of not being able to breathe and rapidly move at the same time was a huge obstacle to colonizing land. The first land tetrapods would have been at a huge disadvantage to even the land arthropods, such as the scorpions, for the vertebrates would have been slow and would have needed to constantly stop to take a breath. This is why we contend that oxygen levels would have been critical: only under high oxygen conditions would the first land vertebrates have had any chance of making a successful living on land.

One consequence of this problem was that the early amphibians and reptiles evolved a three-chambered heart. This kind of heart is found in most modern amphibians and reptiles, and is adaptive for creatures that have the problem of inferior respiration while moving. While a lizard is chasing prey it is not breathing, and thus the fourth chamber of the heart, which would be pumping blood to the lungs, is superfluous. The three chambers are used to pump blood throughout the body, but the price that must be paid is that it takes the lizard longer to reoxygenate the blood when activity ceases.

At this point we can summarize and discuss the variables in land animal reproduction and try to relate these to generalizations about both oxygen levels and temperature. There are two possible strategies, as we have seen: egg laying or live birth. In the egg case, the eggs are either covered with a calcareous shell cover or a softer, more leathery shell cover. Today, all birds utilize calcareous eggs, while all living reptiles that lay eggs use the leathery covering. Unfortunately, there is little information about the relative oxygen diffusion rates for leathery—or parchment—eggs compared to calcareous eggs.