Outcasts of the Islands

1.

In 1964, while studying bird evolution on the tropical Pacific island of New Guinea, I happened to set up camp among a tribe known as the Fore. I soon found my attention drawn away from birds to a human tragedy unfolding around me. Many of the tribespeople, children as well as adults, were limping on crutches, or unable to control their facial muscles, or lying semi-paralyzed in their huts. When I asked what was wrong, their healthy relatives answered with the single word “kuru,” as if no more explanation were needed. Kuru, the Fore way of death, is now internationally notorious as a neurological disease, always fatal, and confined to that one tribe living in a group of mountain valleys only a few hundred square miles in extent.

As I proceeded through the New Guinea highlands in search of birds, every ten or twenty miles I passed into the territory of a different tribe, each with its own language, its own culture—and its own disease or genetic anomaly. The second tribe that I encountered had many albinos; the third, the world’s highest incidence of leprosy; and others had high frequencies of male pseudohermaphroditism, or of a disorder making the skin resemble that of a crocodile, or of misshapen red blood cells. Scientists have learned that each such local condition stems from various combinations of local infectious agents, adaptations, and genes.

While these pathologies may at first seem to be nothing more than exotic diseases confined to faraway peoples, they have proved enormously influential in the development of medical science, for two reasons. First, for physicians, New Guinea’s mountain valleys harbor so many locally distinct human populations that the resulting range of entrenched diseases rivals that of Europe and the United States, with their much larger but more homogeneous human populations. Some of those diseases first recognized in New Guinea turned out to provide decisive insights into more widespread conditions—notably kuru, which has been the best model for understanding Creutzfeldt-Jakob disease, mad cow disease (alias bovine spongiform encephalopathy), and possibly Alzheimer’s disease. Second, for evolutionary biologists, New Guinea’s human populations allow us to study processes of evolution and genetic change under conditions much more relevant to ourselves than the usual animal population studies cited in any textbook of biology.

Two such diseases on other tropical Pacific islands are subjects of the latest book by the distinguished neurologist and author Oliver Sacks. Both diseases have a known or possible genetic basis: achromatopsia (complete color blindness) on Pingelap, and lytico-bodig (a highly variable neurological disorder) on Guam. Dr. Sacks is already deservedly admired for previous successful books (e.g., Awakenings, An Anthropologist on Mars, The Man Who Mistook His Wife for a Hat) distinguished by beautifully written, poignant studies of individuals with neurological diseases, and by Dr. Sacks’s ability to bring himself and his readers into the unusual states of mind of his patients. The current book, based on two visits by Dr. Sacks to Pingelap and Guam and two neighboring islands, is in a similar mold. But The Island of the Colorblind contains much more, because the species of ferns and cycads confined to those islands restimulated Dr. Sacks’s childhood interest in plants, especially since one theory of lytico-bodig attributes the disease to the effects of eating poisonous cycad seeds. Hence Dr. Sacks’s new book is an account of patients with two neurological disorders but also of island plants, islands as laboratories of plant and animal evolution, and many other aspects of islands.

The small atoll of Pingelap lies in the Pacific 180 miles from the volcanic island of Pohnpei, which in turn lies 2,200 miles east of the Philippines and 1,200 miles northeast of New Guinea. Dr. Sacks was drawn to Pingelap because fifty-seven of its seven hundred human inhabitants suffer from a hereditary form of colorblindness termed achromatopsia. About one third of the remaining inhabitants carry one gene for achromatopsia without reporting symptoms. From Pingelap, which he visited with an ophthalmologist friend, Dr. Sacks went on to study a remote village of Pingelap émigrés on Pohnpei.

Our eyes contain four types of light receptors, termed rods and cones. Rods function under conditions of dim light and give us our nighttime white/grey/black world view without color, while the three types of cones function under daylight conditions of bright light and contain three different pigments that respond differently to lights of different colors (different wavelengths). What most so-called colorblind Americans and Europeans suffer is in fact just the lack of one of those three cone types; they are unable to distinguish red from green but have no other impairment of vision or color perception. In contrast, Pingelap achromatopes have no cones at all, only rods, so that they experience a colorless world like that experienced by the rest of us at night. In bright light the rod pigments become bleached out, and cone vision takes over for us “normals”—but not for Pingelap achromatopes. Hence while their condition is often loosely described as complete colorblindness, Dr. Sacks makes clear from his interviews with people on Pingelap that the much more disabling consequence for them is that they are dazzled by daylight and literally blinded by bright sunlight.

Achromatopsia is not confined to Pingelap but occurs throughout the world. For example, Dr. Sacks visited Pingelap in the company of a Norwegian scientist, Knut Nordby, himself an achromatope, who established instant rapport with Pingelap achromatopes because he at last found himself in a society with many people sharing his condition, and they at last saw an outsider coping successfully with a condition identical to theirs. En route home from Pingelap, Dr. Sacks stopped in Berkeley to visit Frances Futterman, an achromatope who has established a worldwide achromatopsia network. But Nordby, Futterman, and other achromatopes outside Pingelap are very rare cases in their own societies.

How did achromatopsia become so disproportionately common on that one tiny, remote island? The probable answer that Dr. Sacks suggests depends on the genetics of achromatopsia and the history of Pingelap. Achromatopsia exemplifies what geneticists term an autosomal recessive disease. Recall that we inherit two copies of most of our genes, one from our father and one from our mother. If you inherit a normal gene for cone light receptors from either of your parents, that one copy suffices for you to make cones in your eyes. Only if you receive a defective mutant gene from both parents and thus have no normal copy of the gene for cone production do you lack the capacity to make cones. But the gene for achromatopsia is very rare everywhere except on Pingelap, and thus it is even rarer to find a marriage in which both the man and the woman happen to have one copy of the achromatopsia gene. In addition, by the laws of genetics—so-called recessive inheritance—only one quarter of the children of such a marriage inherit the defective gene from both parents and end up as achromatopes. That is, throughout most of the world the achromatopsia gene is rare, marriages between two people carrying the genes are much rarer, and achromatopic offspring of such marriages rarer yet.

Pingelap history is unusual in exemplifying what evolutionary biologists describe as the “founder effect.” The flat, low-lying island was completely inundated in 1775 by a typhoon that killed 90 percent of the population outright; most of the immediate survivors were then gradually killed by starvation. The population of a thousand was thus reduced to only about twenty, probably including just about half a dozen adult women in their reproductive years and half a dozen adult men, one of whom happened to be the hereditary chief (titled the nahnmwarki), whom one might expect to have sired more than a merely proportional share of the resulting children. Within a few decades the twenty survivors bred their numbers back to a hundred, but the first achromatopic children were born in the fourth generation after the typhoon, and within a few more generations the frequency of such children rose to its present level of somewhat under 10 percent of the population. Pingelap genealogies suggest that the nahnmwarki himself was the source of the achromatopsia gene transmitted to all current gene-bearers on Pingelap. Probably the gene was formerly present but rare on Pingelap as elsewhere in the world, so that asymptomatic gene carriers—people with just one copy of the gene—rarely happened to marry. Only when the typhoon reduced the adult male population to the gene-carrying nahnmwarki and a few other men of lower status was Pingelap repopulated with a newly founded, inbred population containing many gene carriers, who inevitably often ended up marrying and producing achromatopic children.

The founder effect, which Sacks briefly describes, is a phenomenon of widespread importance in evolutionary biology, because the populations of most animal and plant species are divided into small effective breeding populations or are periodically reduced to small populations by environmental accidents like Pingelap’s typhoon. While evolutionary biologists usually deduce principles from animal studies and apply those principles to humans, the reverse is true for the founder effect, because human beings are unique among animal species in being all individually named and inordinately concerned with recording their genealogies. That makes it possible to trace some locally frequent genetic traits back to one or a few founding ancestral individuals. For instance, hexadactyly—six-fingered dwarfism—is exceptionally common in the Amish population of Lancaster County, Pennsylvania—not because something about Lancaster County makes a sixth finger uniquely useful there, but because the few original founders of the Lancaster Amish population included a certain Mr. Samuel King and his wife, at least one of whom happened to carry the gene for hexadactyly. Again, a genetic condition called osteodental dysplasia, in which all one’s teeth fall out by the age of twenty, is uniquely common in the so-called Cape Colored population of South Africa, not because South Africans don’t need teeth but because a polygamous dysplasic immigrant named Arnold, assisted by his seven wives, propelled the gene to high frequencies in the recently founded population.

It is only through lucky chance that geneticists studying continental populations were able to trace the operation of the founder effect among the Lancaster Amish and the Cape Coloreds. In contrast to populations of remote islands, most continental populations are divided into large political units with extensive genetic homogenization, tending to eliminate any original concentrations of a local gene that are due to founder effects. Such effects for human populations are clearest on islands and on those few parts of the continents not homogenized in the last thirteen thousand years. Until about thirteen thousand years ago, all humans on Earth were divided into small local populations of hunter-gatherers, probably each with its own “private” genes, like Pingelap and its achromatopsia. Since then, continental human populations have undergone three successive waves of genetic homogenization. The first was the expansion of the territory of the earliest farmers at the expense of the hunter-gatherers, erasing much pre-existing gene diversity over wide areas. Examples include the expansion of Bantu farmers over sub-Saharan Africa beginning around five thousand years ago, the expansion of Austronesian farmers over the Philippines and Indonesia beginning around six thousand years ago, and perhaps the expansion of farmers bringing Indo-European languages to Europe around ten thousand years ago.

The second wave of homogeniza-tion began around five thousand years ago with the formation of political states, which promoted intermarriage among a state’s people. The third wave is now the one following airplanes and Coca-Cola across oceans and large land masses, and producing genetic melting pots on Hawaii and elsewhere. That is, human genetic diversity must have been much higher in the past than at present, as new populations were constantly being founded and expanding to carry their private genes over small local areas. Pingelap and New Guinea are thus far more important to geneticists than their tiny fraction of the world’s population would suggest, because they show us our genetic landscape as it used to be. For a long time now, the global trend has been toward mixing populations. But the world’s peoples still have a long way to go before all human populations become coffee-colored and share similar gene frequencies.

Nonetheless, the founder effect acting alone is not the sole possible explanation for achromatopsia’s high frequency on Pingelap. Another, more speculative possibility involves what geneticists term a “balancing selective advantage”: perhaps achromatopsia carries with it some subtle advantage offsetting its obvious disadvantage of impaired vision in bright light. The genetic literature is full of examples of “bad” genes with “good” effects, especially genes for autosomal recessive diseases like achromatopsia.

The best-understood example is the sickle-cell gene common in black Africans and in their African-American descendants. That gene causes our red blood pigment, hemoglobin, to be synthesized in an altered form, resulting in red blood cell oxygen levels below normal. Individuals “homozygous” for the sickle-cell gene—that is, with two copies of it, one inherited from one’s mother and the other from one’s father—produce only sickle-cell hemoglobin and lack normal hemoglobin. Those homozygous individuals end up with low red blood cell oxygen concentrations tending to cause a blood disease, which was often fatal before modern medicine and still can be today. But individuals with only one copy of the sickle-cell gene—so-called heterozygotes, who are much more numerous than the homozygotes—develop only mildly low red blood cell oxygen concentrations, not so low as to make them sick but still low enough to damage malaria parasites during their life-cycle stage within red blood cells.

As a result, in tropical areas of Africa where malaria is the most dangerous infectious disease, sickle-cell heterozygotes have some inherited resistance to malaria and survive even better than do homozygous “normal” individuals producing only normal hemoglobin. The mild selective advantage to the numerous heterozygotes offsets the severe disadvantage to the less numerous sickle-cell homozygotes. The offsetting, of course, is not in the sense of a moral balance whereby some people must die for the good of their relatives, but in the sense of a balancing selection that prevents the sickle-cell gene from being eliminated by natural selection and maintains it in malarial areas of Africa.

Dr. Sacks describes a possible such advantage for achromatopsia under Pingelap conditions, though he abstains from speculating that it could contribute to Pingelap’s high frequency of achromatopsia. He mentions that achromatopic homozygotes on Pingelap are reported as seeing better than normal people under dim-light conditions at dusk and dawn and on moonlit nights. Specifically, the achromatopes are outstandingly successful at night fishing because they can detect fish underwater in dim light. That’s a vital skill for an island population heavily dependent on fish for protein. And as night fishers, Sacks writes, “the achromatopes are preeminent; they seem able to see the fish in their dim course underwater, the glint of moonlight on their outstretched fins as they leap….” Hence I found myself wondering whether heterozygotes carrying the gene for achromatopsia might also have somewhat better night vision than normal people, without the homozygotes’ debilitating impairment of poor day vision. If the heterozygotes thus were somewhat more successful than normal individuals at catching fish and thus attracting marriage partners and feeding their children, they could enjoy an advantage in traditional Pingelap society. In that case, Pingelap achromatopsia would exemplify a founder effect subsequently amplified by natural selection.

Here again isolated little Pingelap may illustrate an evolutionary phenomenon (balancing selection) of worldwide importance, with sickle-cell hemoglobin being only one among myriads of human examples. Dr. Sacks describes one related example, involving the genes that let us store carbohydrate as fat and that may thereby precipitate diabetes in adults. Those fat storage genes are disadvantageous to most readers of these pages—i.e., people whose Western life style includes high sugar intake and low physical activity—because they may thereby end up overweight and diabetic. But fat storage genes could be highly advantageous to people with a spartan life style: if you are especially efficient at storing the occasional high-carbohydrate meal as fat, you will be less likely to starve under the more usual conditions of marginal food supply. Just imagine how Pacific islands must have been originally discovered and populated by long, open-ocean canoe voyages into the unknown. At the end of such a voyage, few of the canoe’s starting passengers might still be alive, and those alive would tend to be the ones who had had the most body fat at the outset. Under the conditions of the traditional vigorous life of Pacific islanders, they would then face a negligible risk of diabetes; but their risk would swiftly increase if they adopted Western habits, a tragedy now being observed among so many Pacific island populations. For instance, on Nauru Island, whose phosphate mines support the most affluent population of Pacific islanders, who shop in supermarkets and are freed from the hard work of farming and fishing, more than half of all adults surviving to the age of sixty end up diabetic.

That balancing of natural selection in adults who may be susceptible to diabetes involves a complex genetic basis. But other genetic diseases familiar by name to readers of this review may involve the balancing of natural selection by the same simple mechanism as that conceivably operating in Pingelap achromatopsia. The cystic fibrosis gene, which is the most common deleterious gene in people of Northern European origins, may have risen to its otherwise puzzling high frequency by protecting heterozygous carriers of the gene against the risk of bacterial diarrheas (formerly the leading killer of infants), while causing the fatal or debilitating condition of cystic fibrosis in homozygotes. The Tay-Sachs gene so common in people of Eastern European Jewish origins may have reached those frequencies by protecting heterozygotes against tuberculosis (formerly a big risk to city-dwellers, as most Eastern European Jews were), even though homozygotes all succumbed in childhood to a fatal disease.

2.

The other neurological disease discussed by Dr. Sacks is lytico-bodig, recently widespread among the native Chamorro population of the Pacific island of Guam. Today, Guam is a melting pot of Filipinos, Spaniards, and other immigrants, but the original inhabitants who greeted Magellan on his voyage of circumnavigation were the Chamorros, a largely Micronesian people.

Lytico-bodig is a doubly puzzling disease. First, its symptoms vary so much that Chamorros call one form lytico and another form bodig, raising some question whether two separate diseases or just one variable disease are involved. The forms range from a paralysis resembling amyotrophic lateral sclerosis (ALS, or Lou Gehrig’s disease) to Parkinsonism and dementia. Second, the cause of lytico-bodig remains debated. Its confinement to Guam’s Chamorropopulation might initially make one suspect yet another genetic disease, like achromatopsia. Indeed, the history of the Chamorro population suggests a founder effect as on Pingelap, because 99 percent of the original population died in the Spanish colonial era from massacres and infectious diseases, leaving only 1 percent to refound the modern Chamorro population. But a genetic theory is difficult to reconcile with the mysterious decline in frequency of new lytico-bodig cases over the last four decades. A competing theory discussed by Sacks involves seeds of wild cycad plants, which are nutritious but contain a substance that has severe toxic effects if it is not carefully removed during preparation of the seeds. Perhaps the recent and now waning lytico-bodig epidemic is a legacy of the Chamorros’ temporary heavy dependence on cycads for food during the Japanese military occupation in World War II. Still other theories invoke a postulated mineral imbalance (too little calcium and magnesium, too much aluminum) or a so-called slow virus.

Whereas Pingelap achromatopsia has attracted little notice from the outside world, medical scientists have devoted much attention to the Chamorros’ lytico-bodig, because it has long been clear that lytico-bodig could be far more than an exotic disease of one remote Third World population. Many features of lytico-bodig are reminiscent of the kuru that I encountered among New Guinea’s Fore people: variable neurological symptoms; confinement to one human population and absence from other populations living in the same environment; tendency of symptoms to run in families; and very long latency for the disease to express itself. Chamorros who leave Guam to settle in California may come down with lytico-bodig forty years later, long after their presumed infection in Guam, if indeed an infectious agent is involved at all. While it remains unknown how the postulated agent was transmitted among Chamorros, we now know that kuru among the Fore is due to a very slow-acting infectious agent formerly transmitted between Fores through endo-cannibalism—the eating of dead relatives. Kuru transmission stopped with the end of Fore cannibalism in 1959, but the consequences to Fore children alive at the time of the last cannibalistic feasts are still emerging today, in the form of new cases of kuru appearing in adults now in their forties. The mysterious waning of cases of lytico-bodig thus analogously suggests the disappearance, around 1952, of some other mode of transmission on Guam.

The medical and scientific insights gained from studying kuru and, potentially, lytico-bodig are enormous. As for the medical insights, at the time that I was working in the Fore area in the 1960s, many medical researchers guessed that kuru was the result of some environmental condition (some toxin?) unique to the Fore area, or possibly was genetic. An infectious agent seemed an unlikely explanation because of kuru’s often very long latency, so much longer than that of most other acknowledged infectious diseases. But Dr. Carleton Gajdusek of the National Institutes of Health received a Nobel Prize for his epochal demonstration that kuru did involve a very slow-acting infectious agent. He inoculated kuru brain extracts into chimpanzees, which proceeded to develop a kuru-like degenerative condition years later. This was a scientific discovery made against great odds: almost any other scientist except Gajdusek would have lost patience and given up long before the chimpanzees developed symptoms.

Kuru (and perhaps lytico-bodig) now proves to be a human analog of long-latency neurodegenerative diseases previously known in animals, including sheep and mink. Creutzfeldt-Jakob disease is another kuru-like human disease that is widespread throughout the world but rare everywhere. Hence the Fore kuru epidemic may well derive from just one original case of Creutzfeldt-Jakob disease, which would have immediately died out elsewhere but was transmitted among the Fore by endo-cannibalism. While readers of this review may dismiss kuru and lytico-bodig as exotic conditions for which they are not at risk, they certainly should be worried about their risk for some other conditions that may be caused by related agents, such as Alzheimer’s disease, multiple sclerosis, and Parkinsonism. Oliver Sacks found that some of the patients he saw reminded him of the victims of postencephalitic Parkinsonism that he described in Awakenings. And many European readers are justifiably terrified by their risk of contracting mad cow disease (BSE, or bovine spongiform encephalopathy), which is caused by an agent very similar to the agent of kuru (and perhaps lytico-bodig).

The BSE agent was transmitted to cows whose feed was partly made up of infected sheep carcasses. In at least fifteen well-attested cases, fatal BSE has now been transmitted to Europeans (mostly British) who ate infected beef. No one can predict whether those cases herald the beginnings of a European epidemic of long-latency BSE, one that will extend over the next forty years, like the kuru and lytico-bodig epidemics.

These are the possible medical implications, but there are equally huge scientific implications. Gajdusek described kuru as a slow virus disease, although he was not able to isolate a responsible virus. He merely made that inference because he knew that the infectious agent was so small that it passed through very fine filters, and the only then-known infectious agents so small were viruses. But growing evidence now suggests that the agent could be a peculiarly folded form of a widespread and normally innocuous protein called a prion, which is found on the surface of some brain cells and which may affect the normally folded protein molecules nearby, causing the formation of destructive clots in the brain. This hypothesis, formulated by Dr. Stanley Prusiner of the University of California at San Francisco, was long rejected as heretical by many other scientists, because it violated some of the most basic accepted tenets of biology.

If the hypothesis does prove to be true, it will in my opinion be the most astonishing discovery by molecular biologists since Watson’s and Crick’s discovery of DNA structure in 1953. Until now we have been assuming that a protein’s three-dimensional structure is uniquely determined by its sequence of amino acid building blocks that supposedly specify how the protein will fold up. But if Prusiner is correct (and his conclusions are still being hotly debated), prion proteins can assume different folded configurations, specified by some factors other than the amino acid sequence itself. Furthermore, while all the genetic information specifying protein structure is supposedly contained in nucleic acids (especially DNA), Prusiner argues, equally heretically, that the form in which protein is folded can also contain such information. It is thus no understatement to say that diseases of those two tropical Pacific island peoples, the Fore and the Chamorros, are revolutionizing our understanding of medicine and biology.

In short, the two medical subjects of Dr. Sacks’s book are important and fascinating. His book also deals with much interesting material about plants, and has short discussions of a wide variety of other engrossing topics. The book contains the vivid and sensitive descriptions we have come to expect from Dr. Sacks’s previous books, especially when he brings us close to the minds and lives of patients with achromatopsia or lytico-bodig. Here, for example, he describes a patient afflicted with bodig:

I was told he had “man-man”—the Chamorro word for staring blankly into space—though this was not a blank staring, a staring at nothing, but an almost painfully engrossed, wistful staring, staring out at the children who played in the road, staring at the occasional passing cars and carts, staring at the neighbours leaving for work each morning, and returning late in the day. Jesus sat on his porch, unblinking, unmoving, motionless as a tortoise, from sunrise till midnight (except on the rare days when high winds or rain lashed across it), forever gazing at a constantly varying spectacle of life before him, an enraptured spectator, no longer able to take part….

We had been told that Jesus might pass the whole day with scarcely a word. And yet he spoke well, even volubly, when we engaged him in conversation; though, it soon became apparent, he waited for our questions. He could respond quite readily, but could not initiate a sentence. Nor, it seemed, a movement either—he might sit totally motionless for hours, unless something or someone called him to move. I was again strongly reminded of my post-encephalitic patients and how they were crucially dependent on the initiative of others, calling them to speech or action.I tore a page out of my notebook, balled it up, and threw the balled paper at Jesus. He had been sitting, seemingly incapable of movement, but now his arm shot up in a flash, and he caught the paper ball precisely. One of his little grandsons was standing by, and his eyes widened with astonishment when he saw this. I continued playing ball, and then asked Jesus to throw the ball at his grandson, and then to another child, and another. Soon we had the entire family playing ball, and akinetic Jesus, no longer akinetic, kept it going between us all. The children had not realized that their “paralyzed” grandfather could move by himself at all, much less that he could catch a ball, aim it accurately, bluff, throw it in different styles and directions, and improvise a fast ball game among them.

In its format, The Island of the Colorblind is unusual, and is best described as a two-tiered mosaic of loosely related chunks. At the top tier, the book consists of five chapters: one organized around Dr. Sacks’s travel to Pingelap, one each around his observations of achromatopes on Pingelap and Pohnpei, one around lytico-bodig on Guam, and one on the island of Rota near Guam. The two diseases are unrelated to each other except for the fact that they afflict peoples of remote Pacific islands; and the chapter on Rota says nothing about diseases but instead describes Rota’s ferns, cycads, and other plants. The book concludes with a fifty-eight-page section of notes, longer than four of the five chapters: ninety-four mini-essays, averaging about one half page each, on diverse other subjects that arose from Dr. Sacks’s trips and that interested him, such as the lives of various botanists, coconut crabs, Darwin’s views about seed dispersal, the effects of golf course construction on Rota, the evolution of bananas, and Micronesian languages.

Each of the five chapters itself is a finer-scaled mosaic, shifting abruptly between diseases, plants, and other subjects. For instance, the thirty-six-page chapter on Pohnpei progresses through the following sequence of topics, devoting slightly under three pages to each: the archaeological ruins at Nan Madol; the human geography of Pohnpei; observations of achromatopic children; the experience of Knut Nordby and his siblings as achromatopes; island bats; medical care on Pohnpei; a description of Pohnpei’s capital; the colonial history of Pohnpei; ethnobotany; forest plants; Dr. Sacks’s experience of becoming stoned on a hallucinogen called sakau; reflections on Pingelap; and the encounter with Frances Futterman in Berkeley that I earlier mentioned.

How well does this patchwork format succeed? My guess is that different readers will react quite differently. For me, the resulting book is frustrating. While the material is intrinsically fascinating, I found my own fascination to derive more from what I already knew about the subjects than from what Dr. Sacks says about them. Many subjects are discussed too briefly for the text to be adequately informative. That problem applies to the treatments even of the two main subjects, achromatopsia and lytico-bodig. Far too little explanation is devoted to the “founder effect,” to autosomal recessive diseases, and to slow viruses or prions to set these two diseases within a larger frame of understanding.

There are also specific problems with using these two diseases as the focus of a book. Dr. Sacks’s main motive for his visit to Pingelap, as reflected in his choice of the book’s title, was to experience an “island of the colorblind.” In Dr. Sacks’s words, he wondered whether Pingelap would be

a culture where the entire concept of color might be missing, but where, instead, other forms of perception, of attention, might be amplified in compensation…, an entire achromatopic culture with its own singular tastes, arts, cooking, and clothing—a culture where the sensorium, the imagination, took quite different forms from our own, and where “color” was so totally devoid of referents or meaning that there were no color names, no color metaphors, no language to express it; but (perhaps) a heightened language for the subtlest variations of texture and tone, all that the rest of us dismiss as “grey.”

Such a society would indeed be interesting. But Pingelap proved not to be such a society. More than 90 percent of its population has normal color vision, and even for the relatively few achromatopes, poor vision in bright light is a more important consequence than lack of a color sense. Lytico-bodig similarly proves to be a somewhat unsatisfying focus: while a slow virus or prion seems to me the most plausible cause, other possible causes have not been ruled out, and the question may ultimately remain unresolvable because the disease is now disappearing spontaneously. Hence the book’s overall shape is a travelogue of two separate trips that interested Dr. Sacks a lot, and that exposed him to many interesting subjects, about many of which he has produced accounts or notes ranging in length from half a page to dozens of pages; but these accounts, although often beautifully observed, are not woven into a coherent, satisfying book.

I should, however, reiterate my admiration for Dr. Sacks’s prose. At a time when it is essential that the public have some understanding of medicine and science, few physicians and scientists are able and willing to explain their subjects well to the general reader. Among those few, Dr. Sacks is pre-eminent in his broad interests, engaging prose, powers of observation, and ability to bring himself and the reader close to another person’s mind. Despite my reservations, many readers will enjoy his new book, just as many listeners enjoy Beethoven’s Fourth Symphony although it is overshadowed by his other symphonies. But Dr. Sacks has set high standards, and by those standards he can do better. I hope that, in his next book, he will use his impressive talents more effectively to produce a more cogently constructed work.