Here's an example of how the reasoning works. We know that humans, chimps, and gorillas are genetically about 98 percent identical and stem from an ancestor (“the Missing Link”) that lived as recently as nine million years ago. Yet those three modern descendants of the Missing Link now exhibit all three types of ovulatory signal: concealed ovulation in humans, slight signals in gorillas, bold advertisement in chimps. Hence only one of those descendants can be like the Missing Link in ovulatory signals, and the other two descendants must have evolved different signals.
In fact, most living species of primitive primates have slight signs of ovulation. Hence the Missing Link may have preserved that condition, and gorillas may have inherited it in turn from the Missing Link (see figure 4.1). Within the last nine million years, though, humans must have evolved concealed ovulation, and chimps must have evolved bold advertisement. Our signals and those of chimps thus di-verged in opposite directions from the cues of our mildly signaling ancestors. To us humans, the swollen derrieres of ovulating chimps look like those of baboons. However, the ancestors of chimps and baboons must have evolved their eye-catching derrieres quite independently, since the ancestors of baboons and of the Missing Link parted company around thirty million years ago.
By similar reasoning, one can infer other points in the primate family tree at which ovulatory signals must have changed. It turns out that switches of signals have evolved at least twenty times. There have been at least three independent origins of bold advertisement (including the example in chimps); at least eight independent origins of concealed ovulation (including its origins in us, in orangutans, and in at least six separate groups of monkeys); and several reappearances of slight signs of ovulation, from either concealed ovulation (as in some howler monkeys) or from bold advertisement (as in many macaques).
In the same way as we've just seen for ovulatory signals, one can also identify points in the primate family tree at which mating systems must have changed. The original system for the common ancestor of all monkeys and apes was probably promiscuous mating. But if we now look at humans and our closest relatives, the chimps and gorillas, we find all three types of mating system represented: harems in gorillas, promiscuity in chimps, and either monogamy or harems in humans (see figure 4.2). Thus, among the three descendants of the Missing Link of nine million years ago, at least two must have changed their mating system. Other evidence suggests that the Missing Link lived in harems, so that gorillas and some human societies may just have retained that mating system. But chimps must have reinvented promiscuity, while many human societies invented monogamy. Again, we see that humans and chimps have evolved oppositely, in mating systems as in ovulatory signals.
Overall, it appears that monogamy has evolved independently at least seven times in higher primates: in us, in gibbons, and in at least five separate groups of monkeys.
Harems must have evolved at least eight times, including in the Missing Link. Chimps and at least two monkeys must have reinvented promiscuity after their recent ancestors had given it up for harems.
Thus, we have reconstructed both the type of mating system and the type of ovulatory signal that probably existed in primates of the remote past, all along the primate family tree. We can now, finally, put both types of information together and ask: what mating system prevailed at each point in our family tree when concealed ovulation evolved?
Here's what one learns. Consider those ancestral species that gave signals of ovulation, and that then went on to lose those signals and evolve concealed ovulation. Only one of those ancestral species was monogamous. In contrast, eight, perhaps as many as eleven, of them were promiscuous or harem-holding species-one of them being the human ancestor that arose from the harem-holding Missing Link. We thus conclude that promiscuity or harems, not monogamy, is the mating system that leads to concealed ovulation (see figure 4.3). This is the conclusion predicted by the many-fathers theory. It doesn't agree with the daddy-at-home theory.
Conversely, we can also ask: what were the ovulatory signals prevailing at each point in our family tree when monogamy evolved? We find that monogamy never evolved in species with bold advertisement of ovulation. Instead, monogamy has usually arisen in species that already had concealed ovulation, and sometimes in species that already had slight ovulatory signals (see figure 4.4). This conclusion agrees with the predictions of the daddy-at-home theory.
How can these two apparently opposite conclusions be reconciled? Recall that Sillen-Tullberg and M0ller found, in step 3 of their analysis, that almost all monogamous primates have concealed ovulation. We now see that that result must have arisen in two steps.
First, concealed ovulation.
Figure 4.3
By combining facts about modern observed species with inferences about ancestral species, one can infer the mating system prevailing when ovulatory signals underwent evolutionary change. We infer that species 3 evolved concealed ovulation from a harem-holding ancestor with slight signs of ovulation, while species I and 2 preserved the ancestral mating system (harems) and slight ovulatory signs.
lation arose, in a promiscuous or harem-holding species. Then, with concealed ovulation already present, the species switched to monogamy (see figure 4.4).
Perhaps by now you're finding our sexual history confusing. We started out with an apparently simple question that deserved a simple answer: why do we hide our ovulations and have recreational sex on any day of the month? Instead of a simple answer, you're being told that the answer is more complex and involves two steps.
What it boils down to is that concealed ovulation has repeatedly changed, and actually reversed, its function during primate evolutionary history. It arose at a time when our ancestors were still promiscuous or living in harems. At such times, concealed ovulation let the ancestral ape-woman distribute her sexual favors to many males, none of which could swear that he was the father of her child but each of which knew that he might be. As a result, none of those potentially murderous males wanted to harm the ape-woman's baby, and some may actually have protected or helped feed it. Once the ape-woman had evolved concealed ovulation for that purpose, she then used it to pick a good caveman, to entice or force him to stay at home with her, and to get him to provide lots of protection or help for her baby-secure in the knowledge that it was his baby too.
On reflection, we shouldn't be surprised at this shift of function for concealed ovulation. Such shifts are very common in evolutionary biology. That's because natural selec-tion doesn't proceed consciously and in a straight line toward a distant perceived goal, in the way that an engineer consciously designs a new product. Instead, a feature that serves one function in an animal begins to serve another function as well, becomes modified as a result, and may even lose the original function. The consequence is frequent reinventions of similar adaptations, and frequent losses, shifts, or even reversals of function, as living things evolve.
One of the most familiar examples involves vertebrate limbs. The fins of ancestral fishes, used for swimming, evolved into the legs of ancestral reptiles, birds, and mammals, which used them for running or hopping on land. The front legs of certain ancestral mammals and reptile-birds subsequently evolved into the wings, used for flying, of bats and modern birds, respectively. Bird wings and mammal legs then evolved independently into the flippers of penguins and whales, respectively, thereby reverting to a swimming function and effectively reinventing the fins of fish. At least three groups of fish descendants independently lost their limbs to become snakes, legless lizards, and the legless amphibians known as cecilians. In essentially the same way, features of reproductive biology-such as concealed ovulation, boldly advertised ovulation, monogamy, harems, and promiscuity-have repeatedly changed function and been transmuted into each other, reinvented, or lost.