Jerison's now classic volume, Evolution of the Brain and Intelligence (1973), drew inspiration from the words of Karl Lashley

Jerison expanded Lashley's views in key respects. First, he postulated that brain size is a natural biological statistic that can be readily used to estimate other anatomical characteristics of the brain including the size of other neural structures, the degree of fissurization, and neuronal density. He recognized, however, one major exception to this rule – the size of the olfactory bulbs appears to have little relationship to that of other brain structures. Second, Jerison articulated the principle of proper mass – ‘the mass of neural tissue controlling a particular function is appropriate to the amount of information processing involved in performing the function.’ (Jerison, 1973, p. 8). Third, Jerison recognized that brain mass varies with body mass, and he developed a model of brain/body size relationships according to which, with increasing body size, brain size expands in a mathematically predictable way in each vertebrate class.

Studies of mammalian and primate brain evolution have traditionally focused on changes in encephalization, that is, changes in brain size statistically adjusted to compensate for changes in body size, rather than on changes in the internal organization of the brain. There are some very sound reasons for stressing size. Mammals do indeed vary dramatically in absolute and relative brain size: at a given body weight, brain weight can vary more than five-fold across species (Stephan et al., 1988). Moreover, brain size changes can have profound consequences for the developmental biology and ecology of mammalian taxa, because larger-brained taxa grow more slowly and live longer than do smaller-brained taxa of comparable body size (Sacher, 1982; Finlay & Darlington, 1995) and because brain tissue is energetically very demanding (Aiello & Wheeler, 1995). Conveniently, brain size is relatively tractable empirically, which is to say that one can measure it with reasonable precision in all sorts of living and extinct taxa, whereas the internal features of brain organization can be examined only with difficulty in extant taxa and not at all in extinct forms. Finally, there can be little doubt that variations in brain size are in some way related to variations in cognitive and behavioral abilities.

But in precisely what ways are brain size, cognition, and behavior related? Harry Jerison has ably articulated the view that encephalization serves as an index of general animal intelligence (see especially Jerison, 1961, 1973), and in doing so has provided the underpinning for modern brain allometry studies.

Although this author's (R.L.H.) disagreements with Harry Jerison are legion (e.g. Holloway, 1966,1974,1979), I have always found his ideas stimulating and thus of great value to my own work regarding human brain evolution. I believe we best honor Harry Jerison by taking his ideas seriously, whether or not we agree with them.

There do not appear to be any serious disagreements that the brain became reorganized as well as enlarged during hominid evolution, but there is considerable controversy as to when reorganization, particularly that relating to the reduction of primary visual striate cortex, Brodmann's area 17, had taken place. (Reviews of these questions can be found in Holloway, 1995, 1996.) Since the only way we will ever know for absolutely certain when this process occurred requires travel with a time machine and some histological sectioning of australopithecine brains, one might wonder why we are writing this paper. It is already apparent from the literature on early hominid brain evolution that a major controversy exists regarding the fossil australopithecine endocasts and their interpretation regarding that infamous landmark, the lunate sulcus. Falk (1983, 1985, 1986) interprets the paleoneurological evidence from the Taung and Hadar (AL162–28) endocasts as indicating that the lunate sulcus was in an anterior pongid-like position. Holloway (1981, 1983, 1984) interprets the evidence as suggesting a posterior, more modern-human like position.

Harry Jerison argued in his Evolution of the Brain and Intelligence that the initial enlargement of the brain of ancestral mammals resulted from refinement of the sensory processes of audition and olfaction, improving adaptation to life in nocturnal niches (Jerison, 1973). Every relationship and transition hypothesized in this statement is an intriguing subject for investigation. Twenty-five years later, challenged by Jerison's vision, we (and he) have learned a great deal about the variation in the structure of the brains of extant mammals and of those mammals represented in the fossil record (Jerison, 1991). Concurrently, the explosion of knowledge in neuroscience has profoundly changed our view of the organization, localization and development of functional systems in the brain. We can now make much better sense of variation in neural structure in terms of behavioral function. In particular, appreciation of how functions can be distributed over multiple brain components is coming to replace our initial single structure-single function models. Furthermore, we can now extend Jerison's initial evolutionary question to ask how brain changes come about developmentally.

In this review, we first examine the extent to which specific structures in the brain may be the targets of selection, or whether change occurs at the level of interconnected functional systems, other organizational units, or at the most global level of brain size. After we have described where variation occurs in the brains of extant mammals, and at which level or levels this variation occurs, our job as developmental neurobiologists is to describe how development produces these differences.

Throughout the lives of most mammalian species, the sense of smell plays an important role in response to chemical messengers that are involved in many different behavioral activities. Pheromones are the most important compounds for olfactory communication. The term pheromone was invented by Karlson & Lüscher in 1959. Pheromones are chemical substances that, when emitted from one animal, cause behavioral or physiological responses in other animals of the same species. Pheromones are secreted by specific organs that are widely scattered on the bodies of different animals. Released pheromones stimulate rapid behavioral changes in the neuroendocrine system and subsequently produce a physiological and behavioral change in the receiving individual. Pheromones also indicate an animal's identity and territory. Pheromones in mammals convey specific information concerning species, gender, physiological phases and identities of animals, thus triggering stereotyped behavioral and neuroendocrine responses. Such responses ensure breeding and hierarchical order in the animal group.

Olfactory communications between conspecific mammals facilitate reproductive processes. Pheromones produced by males and females influence their sexual behavior and hormone activity (Marchlewska-Koj, 1984). In most species, males can distinguish between females in estrus or anestrus phases by their scents. Females, too, are able to identify sexually active males by odor. Production of such olfactory stimulants is controlled by gonadal hormones, mainly testosterone. Pheromones produced by males can accelerate puberty in juvenile females, induce estrus in anestrus females and block pregnancy in recently inseminated females.

It is just over a quarter of a century since Harry Jerison published the first edition of his seminal work, Evolution of the Brain and Intelligence (1973). At the time and in the decades since then, we see it as one of the most significant paleo-neurobiological landmarks which heralded the final quarter of the twentieth century. As we enter the twenty-first century, that work remains ineluctably a signpost for the coming era. It is pleasurable indeed to offer homage to Dr Jerison – who was born one day before me in what seems to have been something of an annus mirabilis, 1925 – and to offer him my thanks for his inspiration and his friendship.

When we consider the ancient hominins, one is struck by the fact of how few have yielded good natural endocranial casts (endocasts).

Modern humans have brains that are between three and five times the size that would be expected for average mammals of human body mass (Aiello & Wheeler, 1995, 1996; Aiello, 1997). Because brain tissue per unit mass has a basal metabolism that is over 22 times higher than the same amount of muscle tissue, a relatively large brain would be expected to have a significant effect on human energy budgets. In the recent literature on human evolution there has been considerable interest in the ways in which the metabolic costs of the large human brain may have either constrained or influenced adaptation and behavior. Focus has centered on how it is possible to grow such a large brain (Martin, 1996), on how adult humans might adjust their energy budgets to maintain their large brains (Aiello & Wheeler, 1995, 1996; Aiello, 1997), and on the implications of the metabolic aspects of brain growth and maintenance for human dietary evolution (Leonard & Robertson, 1992, 1994, 1997), life history evolution (Foley & Lee, 1991), social evolution (Key & Aiello, 1999, 2000), and symbolic evolution (Power & Aiello, 1997). One recent hypothesis has also suggested that the increase in relative brain size during the course of human evolution might be better explained by the metabolic resources available to mothers during gestation and lactation rather than by any specific behavioral feature (for example feeding ecology or complexity of social organization) that might be postulated to exert a selective pressure for a relative increase in brain size (Martin, 1996).

The neocortex is a hallmark of mammalian brain evolution and is the structure that provides the basis of human mental capacity and uniqueness. Since the time of its emergence in a mammalian ancestor perhaps 250 million years ago, the neocortex has expanded in both relative and absolute size independently in several mammalian lineages. This expansion is particularly apparent in anthropoid primates, in which the neocortex comprises up to 80% of the brain mass. The expansion occurs primarily in the surface area rather than in thickness. Further, the neocortex is parcellated into different cytoarchitectonic areas, which increased in number, size and complexity during the cortical evolution.

Traditionally, our insights into the evolution of the neocortex have come from physical anthropology and comparative anatomy (e.g. Ariëns Kappers et al., 1936; Armstrong & Falk, 1982; Butler, 1994; Herrick, 1948; Jerison, 1991; Kaas, 1988; Nauta & Karten, 1970; Northcutt & Kaas, 1995; Preuss, 1993). In contrast, genetic, molecular and cellular mechanisms by which the cerebral cortex might have evolved are only beginning to be scientifically explored (e.g. Simeone, 1998; Smith Fernandez et al., 1998). The present review is an attempt to interpret some of the recent advances in neuroembryology within the context of neocortical evolution. That the embryonic development of living species can provide clues about possible mechanisms underlying evolution is a well-established approach in evolutionary biology that has been used extensively (e.g. Gerhart & Kirschner, 1997; Gould, 1977; Haeckel, 1879; Richardson et al., 1997; Striedter, 1997, 1998).

Although it has long been recognized that, around the world, adult men have larger brains on average than adult women (Pakkenberg & Voigt, 1964; Pakkenberg & Gundersen, 1997), many workers have traditionally viewed men's larger brains as simple correlates of their larger mean body masses. Other findings which suggest that the internal structure of the brains of men and women are, on average, organized differently (summarized in Kimura, 1992), and that the two sexes perform differently on certain cognitive tasks (Kimura, 1992; Falk, 1997) have traditionally been minimized with the latter being attributed largely to variations in developmental experience, as noted by Kimura (1992). Recent reports in the neurosciences, however, underscore the differences between the brains of men and women in gross volume adjusted for body size (Ankney, 1992; Falk et al., 1999), and in internal anatomy that reflects neurological wiring (Gur et al., 1999; Giedd et al., 1996b). Furthermore, convincing arguments are emerging which support the hypothesis that the neuroanatomical differences between the sexes form the substrates for their differences in average cognitive processing (Andreasen et al., 1993; Gur et al., 1999). The purpose of this chapter is to outline some of these new findings and to interpret them within an evolutionary framework.

Sex differences in brain size at equivalent body masses

Humans

Ankney (1992) plotted bivariate regression equations that relate brain weight to body height and body surface area for men and women from data provided in the literature (Ho et al., 1980a, b).

Nervous systems

One of the characteristics of higher animals is their possession of a more or less elaborate system for the rapid transfer of information through the body in the form of electrical signals, or nervous impulses. At the bottom of the evolutionary scale, the nervous system of some primitive invertebrates consists simply of an interconnected network of undifferentiated nerve cells. The next step in complexity is the division of the system into sensory nerves responsible for gathering incoming information, and motor nerves responsible for bringing about an appropriate response. The nerve cell bodies are grouped together to form ganglia. Specialized receptor organs are developed to detect every kind of change in the external and internal environment; and likewise there are various types of effector organ formed by muscles and glands, to which the outgoing instructions are channelled. In invertebrates, the ganglia which serve to link the inputs and outputs remain to some extent anatomically separate, but in vertebrates the bulk of the nerve cell bodies are collected together in the central nervous system. The peripheral nervous system thus consists of afferent sensory nerves conveying information to the central nervous system, and efferent motor nerves conveying instructions from it. Within the central nervous system, the different pathways are connected up by large numbers of interneurons which have an integrative function.

In the previous chapter we examined some of the properties of skeletal muscles without giving much consideration to the mechanisms of the contraction process. It is as if we had investigated the properties of a motor vehicle by measuring its top speed, its fuel consumption, and so on, without finding out how the engine works. Now it is time to look under the bonnet.

Excitation–contraction coupling

The way in which the muscle cell is excited has been described in Chapter 7: an all-or-nothing action potential sweeps along the whole length of the fibre. This is followed by contraction, and the process linking the two events is called the excitation–contraction coupling process. The question we have to answer is: how does the action potential cause contraction?

Depolarization of the cell membrane

When muscle fibres are immersed in a solution containing a high concentration of potassium ions, they undergo a relatively prolonged contraction called a potassium contracture. The tension produced is related to the potassium concentration in a sigmoidal way as is shown in Fig. 10.1. The membrane potential is of course reduced under these conditions (see Chapter 3), so it seems that depolarization is an adequate stimulus for contraction. Normally this depolarization occurs during the propagated action potential.

Skeletal muscles are the engines of the body. They account for over a quarter of its weight and the major part of its energy expenditure. They are attached to the bones of the skeleton and so serve to produce movements or exert forces. Hence they are used in such activities as locomotion, maintenance of posture, breathing, eating, directing the gaze and producing gestures and facial expressions.

Skeletal muscles are activated by motoneurons as we have seen in previous chapters. Their cells are elongate and multinuclear and the contractile material within them shows cross-striations, hence skeletal muscle is a form of striated muscle. In contrast, cardiac and smooth muscles have cells with single nuclei, and smooth muscles are not striated; we shall examine their properties in a later chapter.

Anatomy

Skeletal muscle fibres are multinucleate cells (Fig. 9.1) formed by the fusion of numbers of elongated uninucleate cells called myoblasts. Mature fibres may be as long as the muscle of which they form part, and 10 to 100 μm in diameter. The nuclei are arranged around the edge of the fibre. Most of the interior of the fibre consists of the protein filaments which constitute the contractile apparatus, grouped together in bundles called myofibrils. The myofibrils are surrounded by cytoplasm (or sarcoplasm), which also contains mitochondria, the internal membrane systems of the sarcoplasmic reticulum and the T system, and a fuel store in the form of glycogen granules and sometimes a few fat droplets.