(Pages 276-279 from Shadows of Forgotten Ancestors

by Carl Sagan and Ann Druyan.)

In recent years it has been possible to go much further, to the very heart of life, to the Holy of Holies, and compare, nucleotide by nucleotide, the DNA molecules of two animals. We can now quantify the kinship of different species. We are able to establish molecular pedigrees, DNA genealogies, which provide the most powerful and compelling evidence that evolution has occurred, as well as tantalizing clues on its mode and tempo. The new tools of molecular biology have yielded insights wholly unavailable to previous generations.

Every animal with a backbone has a bloodstream in which hemoglobin is the oxygen carrier. Hemoglobin is composed of four different kinds of protein chains wrapped about one another. One of them is called beta-globin. A particular region of the ACGT sequence codes for beta-globin in all these animals, but only about 5 percent of the region is occupied by the actual instructions for this protein chain. Much of the remaining 95 percent are nonsense sequences - so here mutations can accumulate without being winnowed out by selection.

When the beta-globin regions of the DNA are compared across the primate order, humans are found to be more closely related to chimps than to anyone else. (The human-gorilla connection comes in a close second.) A new basis for our chimp connection is uncovered: not just the bones, the organs, and the brains, but also the genes - the very instructions for making chimps and humans - are almost indistinguishable.

The DNA sequence that codes for beta-globin is roughly fifty thousand nucleotides long; that is, along a given strand of the DNA molecule, fifty thousand As, Cs, Gs, and Ts in a particular sequence describe precisely how to manufacture the beta-globin of the species in question. If the sequences of humans and chimpanzees are compared nucleotide by nucleotide, they differ by only 1.7%. Humans and gorillas differ by 1.8%, almost as little; humans and orangutans, 3.3%; humans and gibbons, 4.3%; humans and rhesus monkeys, 7%; humans and lemurs, 12.6%. The more the sequences of two animals differ, the more remote (both in relatedness and, usually, in time) is their last common ancestor.

When ACGT sequences that are mainly active genes are examined, a 99.6% identity is found between human and chimp. At the level of the working genes, only about 0.4% of the DNA of humans is different from the DNA of chimps.

Another method is first to take the DNA from a human being, unzip the double helix, and separate the two strands. Then do the same for a comparable DNA molecule of some other animal. Put the two strands together and let them link Up. You've now made a "hybrid" molecule of DNA. Where the complementary sequences are closely the same, the two molecules will tightly bind to each other, forming part of a new double helix. But where the DNA molecules from the two animals differ a great deal, the bonding between the strands will be intermittent and weak, and whole sections of the double helix will be flopping loosely. Now take these hybrid DNA molecules and put them in a centrifuge; spin them up so the centrifugal forces tear the two strands apart. The more similar the ACGT sequences are - that is, the more closely related the two DNA strands are - the more difficult it will be to tear them apart. This method does not rely on selected sequences of DNA information (that coding for beta-globin, for example) but on vast amounts of hereditary material, making up whole chromosomes. The two methods - determining the ACGT sequences of selected portions of DNA, and DNA hybridization studies - give remarkable overall agreement. The evidence that humans are most closely related to the African apes is overwhelming.

On the basis of all the evidence, the closest relative of the human proves to be the chimp. The closest relative of the chimp is the human. Not orangs, but people. Us. Chimps and humans are nearer kin than are chimps and gorillas or any other kinds of ape not of the same species. Gorillas are the next closest relatives, both to chimps and humans. The more remote the kinship - when we go to monkeys or lemurs or, say, tree shrews - the less the similarity in sequence. By these standards, humans and chimps are about as closely related as horses and donkeys, and are closer relatives than mice and rats, or turkeys and chickens, or camels and llamas.

"All right," you might say, "maybe chimp anatomy is almost the same as mine. Maybe the chimp's cytochrome c and hemoglobin are almost the same as mine. But the chimp isn't nearly as smart as I am, as well-organized, as hardworking, as loving, as moral, as devout. Maybe when the genes for these traits are discovered, bigger differences will be found." Yes. Maybe you're right. And even that 99.6% identity can be misleading. A 0.4% difference is substantial, because the DNA in any cell in either species is composed of some 4 billion ACGT nucleotides; of them conservatively 1% are in working, no-nonsense portions of the DNA and constitute the genes as such.

The number of operational ACGT nucleotide pairs that are different between humans and chimpanzees must then be roughly 0.4% times 1% times 4 billion, or 160,000. If these are the working parts of genes each 1,000 nucleotides long, each of which codes for a separate enzyme, then the number of completely different kinds of enzymes that humans have and chimps don't, or vice versa, would be somewhere around 160,000/l,000 or 160. We recall that enzymes have a powerful leverage; they preside over changes in the chemistry of the cell, which can happen very fast; one enzyme can process a multitude of molecules. A hundred enzymes, if they're the right enzymes, might make a very big difference. A hundred enzymes seems more than enough to account for Huxley's metaphorical description of the difference between apes and humans: "a hair in the balance-wheel, a little rust on a pinion, a bend in a tooth of the escapement, a something so slight that only the practiced eye of the watchmaker can discover it." Some enzymes would affect estrus, some stature, some fur, some climbing and leaping abilities, some development of the mouth and larynx, some changes in posture, toes, and gait. Many of them would be for a bigger brain with a bigger cerebral cortex, and new ways of thinking beyond the reach of apes.

What's more, a hundred enzymes changed is certainly an underestimate. Probably none of the differences between chimps and humans requires entirely new enzymes to be evolved. A small number of changes, maybe only a change in a single nucleotide, is adequate to render an enzyme inoperable or to change its function. And many of the differences may not be in the genes themselves, but in the promoters and enhancers, the regulatory elements of the DNA that control when and for how long certain genes should be operational. So even a 0.4% difference could, for all we know, imply profound differences in certain characteristics.

Still, chimps are nearer relatives to us than any other animal on Earth. A typical difference between your DNA - all of it, including the untranscribed nonsense - and that of any other human being is roughly 0.1% or less. By this standard, chimps differ from humans only about 20 times more than we differ from one another. That seems awfully close. We must be very careful that those "mortifying reflections" of which Congreve spoke do not make us exaggerate the differences and blind us to our kinship. If we want to understand ourselves by closely examining other beings, chimps are a good place to start....

(Page 343 from Shadows of Forgotten Ancestors by Carl Sagan and Ann Druyan.)

We humans have not evolved from any of the two hundred other primate species alive today; rather, we and they have evolved together from a succession of common ancestors. As we reconstruct the primate family tree, we discover who our closest relatives are. The behavior of the primates varies so widely, even between species in the same genus, that it really does make a difference for our view of ourselves which ones are our nearest relatives.

The answer, as we've already described, seems to be that the chimps are our closest kin, sharing some 99.6% of our active genes. We know from DNA sequencing, as you would of course suspect, that bonobos and ordinary chimps are a lot more like each other than either of them are like us. But 99.6% is very close. We must share many characteristics of both. (Indeed, there must be behavioral traits that we share with our most distant primate cousins.)

By using molecular and anatomical evidence, together with the record in the rocks, the entire family tree of primates can be drawn, at least approximately, and a timeline placed upon it. The evidence from the bones and from the molecules are not in perfect accord, although they are beginning to converge; in this book, we have given weight to gene sequencing and DNA hybridization data. According to the molecular evidence, gorillas branched off from the evolutionary line leading to us about 8 million years ago; the still unidentified, now-extinct common ancestor of humans and chimpanzees separated from the gorillas maybe a million years later. Very quickly thereafter, the lines to chimps and humans began evolving toward their separate destinies. On a planet that's been inhabited a thousand times longer, that's pretty recently, as recent as the last two weeks in the life of a fifty-year-old human. This doesn't mean that humans and chimps themselves began 6 million years ago; only that our common twig in the evolutionary tree branched out then....