further reading


Physiology papers

J.M. Diamond.  (1964)  The mechanism of isotonic water transport.  Journal of general physiology 48: 15-42.

Many organs of our bodies consist of cavities lined by sheets of cells called epithelia, whose major functions include transporting water and dissolved salts and nutrients out of or into the cavity.  Important epithelial organs include the kidneys, intestines, and stomach, but these are anatomically and physiologically complex.  I initially established my reputation in membrane physiology by studies of the gall bladder, a simple and easily studied epithelium that shares some physiological processes with more complex and hard-to-study organs.  In this paper I asked how salt transport by the gall bladder also resulted in water transport such that the transported salt solution is “isotonic” – i.e., has the same osmotic pressure as the animal’s body fluids.  I removed a gall bladder from an animal, suspended it in moist air, filled it with salt solutions whose osmotic pressure I varied, and measured the osmotic pressure of the transported fluid dripping off the outside of the gall bladder.  (Normally, when the gall bladder is in the animal, the transported fluid is absorbed into the animal’s veins and isn’t available for direct measurements).  I found that the transported fluid always had the same osmotic pressure as the solution filling the gall bladder, thereby proving that transported salt pulls water along by osmosis.

J.M. Diamond and w.H. Bossert.  (1967)  Standing-Gradient osmotic flow: a mechanism for solute-linked water transport in epithelia.  journal of general physiology 50: 2061-2083.

This theoretical paper followed my experimental discovery of the previous paper, by asking how transported salt could pull enough water by osmosis to achieve complete osmotic equilibrium, and didn’t just diffuse away.  In collaboration with an electron microscopist, John Tormey, John and I discovered that water transport by the gall bladder traversed long and narrow channels between cells.  That suggested to me a hypothesis: perhaps the shape of the channels could explain complete osmotic equilibration, if salt were transported into the channels, pulled water into the channels by osmosis, and the water flushed the salt out of the channels before the salt could diffuse away without water.  Did this hypothesis make quantitative sense?  I was able to derive a differential equation describing osmotic equilibration in long narrow channels.  But I couldn’t solve or compute the equation, so I collaborated with Bill Bossert, a friend who is a mathematical biologist and did know how to compute the equation’s solutions.  Our calculation showed that my hypothesis was indeed quantitatively plausible. 

J.M. Diamond and E.M. Wright.  (1969)  Biological MEmbranes: the physical Basis of ion and non-electrolyte selectivity.  Annual Reviews of PHysiology 31: 581-646.

A striking feature of biological membranes is that solutes (dissolved molecules) that are very similar to each other in size and physical properties often pass through (permeate) a membrane at very different rates.  That’s true both of ions (electrically charged solutes) such as sodium and potassium, and non-electrolytes (uncharged solutes) such as sugars and alcohols.  My colleague Ernest Wright and I had been comparing the permeability of gall bladder membranes to ions such as sodium and potassium and chloride.  We had also devised an electrical method of quickly measuring gall bladder permeability to non-electrolytes, and we had used that method to measure gall bladder permeability to over 200 different non-electrolytes.  In this review article we extended our gall bladder experience to discuss the physical principles underlying discrimination between similar ions and similar non-electrolytes by biological membranes in general.  In the process we devised new methods for quantitatively calculating the effects of different pieces of a non-electrolyte molecule on permeability.

E. FrÖmter and J.M. Diamond.  (1972)  Route of Passive Ion Permeation in Epithelia.  Nature New Biology 235: 9-13. 

J.M. Diamond.  (1974)  Tight and Leaky junctions in epithelia: A perspective on kisses in the Dark.  Federation Proceedings 33: 2220-2224.

J.M. Diamond. (1977)  The epithelial junction: bridge, gate, and fence. [1976 Bowditch Lecture, American Physiological Society]. The Physiologist 20: 10-18.

Epithelia consist of sheets of cells joined by structures that are visible in the electron microscope and that had been termed “tight junctions” by anatomists.  Behind this name lay the tacit assumption that the junctions really were tight, i.e. that they didn’t permit the passage of water and salts and other small molecules.  Eberhard Frömter and I tested this assumption on the gall bladder by an electrical method (cable analysis), taking advantage of the fact that salt’s components (termed ions) are electrically charged and carry electrical currents.  (In contrast, electrical currents in the wires of our machines and our buildings are carried by electrons, not by ions).  Eberhard and I found that the gall bladder’s tight junctions, far from being tight, are actually leaky: they are the main route by which ions cross the gall bladder.  In some other epithelia, though, the tight junctions really are tight.  This discovery turned out to explain other differences among epithelia as well, especially epithelial differences in electrical resistance and in osmotic pressure of the transported fluid.

Ecology Papers

J.M. Diamond. (1972)  Biogeographic kinetics: estimation of relaxation times for avifaunas of southwest Pacific islands. Proceedings of the National Academy of Sciences, USA 69: 3199-3203.

Isolated populations of species, such as island populations, risk going extinct with a probability-per-unit-time that is higher for smaller populations.  This risk of extinction arises as a practical concern for national parks, which in effect are islands of protected habitat in a sea of man-modified habitat.  For species confined to habitats protected in national parks, how great is the risk of extinction in a national park of some specified area?  I took advantage of a natural experiment to obtain estimates of extinction rates.  In the shallow waters surrounding New Guinea are numerous islands that were connected to New Guinea at Ice Age times of low sea level, and that became islands only when Ice Age glaciers melted and sea level rose around 10,000 years ago.  Those islands now harbor many fewer bird species than the nearly full suite of New Guinea species that they must have supported when they were pieces of New Guinea until 10,000 years ago.  I thereby calculated extinction rates for an island’s whole avifauna, and also separately for just those bird species that don’t fly across water and surely became isolated when sea level rose.  It turned out that islands with an area of a few hundred to a few thousand square miles have retained about half of their original bird populations after 10,000 years, but that small islands of less than one square mile have lost all their original populations and ended up supporting only those bird species capable of recolonizing the island overwater.

J.M. Diamond. (1973)  Distributional ecology of New Guinea birds. Science 179: 759-769.

After I began field work on New Guinea birds in 1964, most of my publications, including a book, were technical accounts describing all the bird species that I encountered in some particular part of New Guinea.  Those accounts were of interest primarily to a few specialists on New Guinea birds.  My 1973 Science paper summarized my findings of broad interest.  I discussed what determines the number of species on an island, why many species are distributed patchily rather than continuously across the landscape, how related species managed to co-exist with each other, how new species arise in the mountains of New Guinea, and how competition affects the behavior and abundance of bird species.  This was also the first publication in which I discussed what insights studies of birds might offer into human evolution and ecology.

J.M. Diamond. (1974)  Colonization of exploded volcanic islands by birds: the supertramp strategy. Science 183: 803-806.

In 1972 I surveyed the birds of several volcanic islands near New Guinea that, like the famous Indonesian island of Krakatoa, had been defaunated by a massive volcanic eruption a few centuries ago.  These islands now support trees, birds, and insects that must be descended from colonists that flew, were blown or carried, or floated to the islands after they had been defaunated.  I found that only a small set of New Guinea birds has become established on these islands, and that especially abundant were some species for which I coined the term supertramps.  These species are specialized for high rates of overwater dispersal and rapid reproduction at the expense of competitive ability.  They thereby become confined to small or recently defaunated islands, because they are among the first species to recolonize those islands, but they are competitively excluded from species-rich islands.

J.M. Diamond. (1975)  The island dilemma: Lessons of modern biogeographic studies for the design of nature preserves. Biological Conservation 7: 129-146.

This paper expanded on a section of my 1972 paper “Biogeographic kinetics,” by explicitly discussing the lessons that studies of island populations can teach us about the conservation of threatened species in national parks and nature reserves.  I described how quickly isolated populations are expected to go extinct in isolated reserves, which species are most likely to go extinct, and how natural reserve systems can be designed in order to minimize extinction rates.  This study and other biological studies of isolated natural populations are now routinely taken into account in conservation planning.

J.M. Diamond. (1982)  Rediscovery of the Yellow-fronted Gardener Bowerbird. Science 216: 431-434.

My most publicized biological discovery, which made the front page of the New York Times, was the rediscovery of New Guinea’s long-lost Yellow-fronted Gardner Bowerbird.  Many of New Guinea’s most spectacular bird species, its birds of paradise and bowerbirds, became known to Western scientists not through European collectors going out to New Guinea and collecting birds at a known locality, but instead by New Guinea and Malay hunters collecting birds and selling them to plume merchants, who in turn sold them to hat shops and feather merchants in Paris and other European cities.  Museum ornithologists described new species that they found in the hat-shop inventories, without knowing exactly where (presumably in New Guinea) the species lived.  Twentieth-century collecting expeditions then tracked down most of the hat-shop birds to their home grounds in different parts of New Guinea, until by 1979 there remained only two of the hat-shop birds whose home grounds were still unidentified.  In 1979 and 1981 I was dropped by helicopter into Indonesian New Guinea’s remote and uninhabited Foja Mountains, where the long-lost Yellow-fronted Gardner Bowerbird (previously known only from four hat-shop specimens) was one of the first birds that I encountered on entering the forest.  I discovered the mating display of the male bowerbird, which turned out to woo females by building a tower of sticks, decorating the bower’s base with blue berries, holding a blue berry against the background of its golden-orange crest, and pointing it towards a female visiting the bower.  In the Foja Mountains I also saw females of a six-wired bird of paradise which I suspected might be the other missing hat-shop bird, Berlepsch’s Bird of Paradise; Bruce Beehler succeeded in confirming this suspicion and observing males in 2005 – 2008.

J.M. Diamond. (1983) Melampitta gigantea: possible relation between feather structure and underground roosting habits. Condor 85: 89-91.

Quite a few New Guinea bird species that were initially known to Western scientists only by collected specimens have turned out to have unusual habits when they became known as living birds.  In addition to the bowerbirds described in two others of my papers posted on this website, these birds with unusual habits include the world’s first known poisonous bird; two bird species whose flesh develops a stink very quickly after the bird dies; a bird that may catch insects by secreting on its palate a sticky substance with an attractive smell and sitting with its mouth open; and birds of paradise whose males woo females by hanging upside-down from a branch and quivering.  Another such case emerged from my discovery of the world first known underground bird.  In 1981 I found that the Greater Melampitta, a large black bird previously known only by six specimens collected without any field observations, roosts inside deep narrow vertical sink holes in karst terrain.  The shafts of the wing and tail feathers are stiff, and the feather vanes (the fluffy part of the feathers) become worn down, possibly because the bird uses its wings and tail to prop itself on the rock walls of vertical sink holes as it ascends from its underground roosts.

J.M. Diamond. (1986) Animal art: variation in bower decorating style among male bowerbirds Amblyornis inornatus. Proceedings of the National Academy of Sciences, USA 83: 3042-3046.

Males of bowerbirds, a family of birds confined to New Guinea and Australia, woo females by building the most elaborately decorated structures erected by any animal species other than humans.  Different bowerbird species and populations prefer to decorate their bowers with natural objects (such as fruits, flowers, stones, and butterfly wings) or human-made objects (such as plastic toothbrushes, car keys, and (formerly) discarded yellow Kodak film cartons) of specific colors.  Hence when in 1983 I visited areas of New Guinea where I expected to encounter bowerbirds, my wife gave me a set of numbered poker chips of seven different colors that I used for choice experiments.  I found that Kumawa bowerbirds decorated their bowers only with black, brown, and gray natural objects, while Wandamen bowerbirds belonging to the same species decorated with objects of many different colors (red, orange, yellow, green, blue, and purple).  Correspondingly, when I decorated the bowers myself with color poker chips, the Kumawa bower owners threw away all of my chips, while the Wandamen birds used my blue, purple, orange, and red chips.  Wandamen bowerbirds stole colored poker chips of preferred colors from each others’ bowers, as I was able to confirm by tracking the movements of my numbered chips among bowers.  Because Kumawa and Wandamen bowerbirds belong to the same species, and because bowers only a few miles apart in the Kumawa Mountains are decorated with different colors, these color preferences varying among and within population probably are not genetically determined but instead reflect local cultural preferences acquired and transmitted by observation and learning – analogous to variation in art and musical preferences among human populations.

J.M. Diamond. (1988) Factors controlling species diversity: overview and synthesis. Annals of the Missouri Botanical Garden 75: 117-129.

One of the central problems of ecology is to understand why some habitats (e.g., equatorial habitats, lowland habitats, wet habitats, and large islands) tend to harbor more bird species than do other habitats (e.g., polar habitats, high montane habitats, arid habitats, and small islands).  The postulated explanatory factors are usually put forward as a non-organized laundry list of factors.  In this paper I proposed a 4-fold grouping of factors that I termed the QQID approach: Q = factors involving resource quality and determining the “number of niches”; Q = factors involving resource quantity plus the quantity of consumers dividing the resources; I = species interaction; and D = dynamic processes such as extinction, immigration, and speciation. 

J. Sanderson, J. Diamond, and S. Pimm. (2009) Pairwise co-existence of Bismarck and Solomon landbird species. Evolutionary Ecology Research 11: 1-16.

How do interactions between species affect the distributions of species?  In 1975 I analyzed the distributions of 150 bird species on 50 islands of the Bismarck Archipelago, lying east of New Guinea in the Southwest Pacific Ocean.  I found that some pairs of ecologically similar bird species never shared the same island but instead replaced each other in a geographically irregular checkerboard pattern.  There were also more complex patterns of exclusion: within some sets of ecological similar bird species, species occurred in only certain permitted combinations (species pairs, triplets, and quartets), out of all the possible combinations that one could write on paper.  I termed these patterns “assembly rules,” which I interpreted as arising partly from competitive exclusion and resource use.  Critics objected that such patterns could also arise merely by chance, and that I had not adequately excluded that possibility.  The result was a vigorous debate in which Michael Gilpin and I on the one hand, and our critics on the other hand, compared actual distributions with those expected by chance, by constructing so-called null matrices (randomly reshuffled distributions of the same species on the same islands).  We or our critics variously concluded that the actual distributions respectively did or didn’t differ from those expected by chance.  However, the modest power of computers available at the height of the debate in the 1970’s and 1980’s limited the number of null matrices that could be constructed, and thus left it uncertain whether the small number of null matrices used was an adequate sample of all possible null matrices.  With the more powerful computers now available, Jim Sanderson and Stuart Pimm and I reexamined this controversy by constructing one million distinct null matrices and showing that they were indeed an adequate sample.  By then, further field work had also generated a larger database: 191 bird species on 183 islands.  With this sample, we showed that more species pairs have mutually exclusive distributions, and fewer species pairs have coincident distributions, than expected by chance.  The proportion of species pairs with exclusive distributions increases with the degree of taxonomic relatedness between the members of the pair.  The exclusive island distributions are largely confined to bird genera in which species segregate from each other spatially (i.e., by altitude, habitat, or island type), rather than genera in which species overlap spatially and segregate non-spatially (i.e., by diet, body size, or foraging technique).  These findings are as one would expect if the exclusive distributions are caused by competition.  We believe that this new analysis resolves the long-standing debate.

Nature articles

Nature Magazine is a leading international weekly science journal that publishes papers in all fields of science.  Most of those papers are written in a technical style and are largely incomprehensible to would-be readers other than specialists in that narrow field of science.  Hence each issue of Nature begins with a News and Views section of half-a-dozen short papers, in which scientists attempt to describe new scientific discoveries in terms that other scientists in other fields, and hopefully also members of the educated public, can understand.  I’ve published 135 of these News and Views papers over the course of the last 35 years, in many fields that include animal behavior, anthropology, archaeology, Bach’s music, conservation biology, ecology, genetics, linguistics, ornithology, physiology, and others.  I post here on this website two of these papers.  One explores the insights that herring gulls offer into two related questions: when, theoretically, should, and when, actually, do, animals practice extramarital sex?  The other paper explores the corresponding theoretical and practical questions for animal divorce, drawing on studies of oystercatchers, black petrels, and kittiwakes.  The latter but not the former paper briefly considers extensions of these bird studies to human behavior.

J.M. Diamond. (1984) Theory and practice of extramarital sex. Nature 312: 196.

J.M. Diamond. (1987) A Darwinian theory of divorce. Nature 329: 765-766.

Magazine articles and op-ed pieces

J.M. Diamond. (1983) Why animals run on legs, not on wheels. Discover 4, no. 9, 64-67.

This was my first article for the general public in a popular science magazine.  Animals and plants evolved many mechanisms and processes that humans separately invented for our own devices and machines.  Those parallel outcomes of biological evolution and invention include jet propulsion, flight, armor, attack by projectiles, illumination, sonar, and much else.  Why, then, did animals not also evolve wheels, which are among the most useful of human inventions?

J.M. Diamond. (1987)  Soft sciences are often harder than hard sciences. Discover 8, no. 8, 34-39.

Laboratory scientists mistakenly believe that the scientific method consists mainly of performing replicated controlled experiments.  That is indeed the preferred method in the so-called hard sciences, which include molecular biology, chemistry, and physics.  In the soft sciences, which include large areas of field biology and of the social sciences, it’s usually impossible, illegal, or immoral to carry out replicated controlled experiments: one has to result to other methods.  For instance, when I study what causes one New Guinea bird species to live at a higher altitude than a related New Guinea bird species, I could in theory quickly settle the issue by experimentally exterminating or trans-locating bird populations and observing what happens to related bird populations – but that experiment would be illegal and immoral, and difficult as well.  Many hard scientists unjustly scorn the soft sciences despite the great importance of the soft sciences for human well-being, and despite the ingenuity that soft scientists must use to resolve the problems of their fields.  When a mathematician unjustly (in my view) convinced our National Academy of Sciences to overturn the election of a distinguished social scientist to Academy membership, I wrote this article about research in soft sciences. 

J.M. Diamond. (1989)  This-fellow frog, name belong-him dakwoNatural History 98, no. 4,16-23.

New Guineans don’t use the English or Latin names for their local bird species.  But every New Guinea group with which I’ve worked has hundreds of names in their local language for the species of birds, plants, spiders, rats, and other creatures around them.  Whenever I arrive in a new New Guinea language area where I haven’t worked before, I put much effort into learning these names for bird species, so that I can discuss local birds with the area’s people and tap their encyclopedic knowledge of natural history.  This article relates how I became aware of these local names, and what I learned about them.

J.M. Diamond. (1989)  How cats survive falls from New York skyscrapers. Natural History 98, no. 8, 20-26.

How do cats survive falls that would unfailingly kill a person, such as a fall from the 32nd floor of a skyscraper?  Why is a cat more likely to be killed by a fall from the 4th floor than from the 32nd floor?  The answers depend on fascinating facts of biomechanics.

J.M. Diamond. (2008)  What’s your consumption factor? New York Times Jan 2, A19.

Several decades ago, it was widely believed that human population growth was the most serious threat to our future.  We now realize that the real problem isn’t population by itself, but consumption.  Ten billion people on Earth would be no problem if they were all in cold storage, not metabolizing or eating or consuming.  Consumption rates vary about 32-fold around the world: each American or Western European or Japanese consumes about 32 times more oil, water, and metal than does each poor African.  Westerners tend to worry about the high rates of population growth in Kenya (population over 30,000,000), but the 7,000,000 inhabitants of Switzerland or of Massachusetts constitute a much bigger burden on world resources, because each of them consumes far more resources than does each Kenyan.  My op-ed piece in the New York Times explores the importance of consumption factors.

J.M. Diamond. (2012)  What makes countries rich or poor? The New York Review of Books. 59 (10): 70-75.

The world’s richest countries, such as Norway and Luxembourg, have average incomes hundreds of times higher than those of the world’s poorest countries, such as Burundi and Sierra Leone.  Why?  Variation among countries in prosperity is one of the central questions of economics.  It’s a question of practical as well as of intellectual interest: if we understood why rich countries are rich, perhaps we could use that knowledge to help poor countries become rich.  A recent book by two distinguished economists, emphasizing the role of institutional factors, stimulated me to write a book review that can serve as an overview of those and other major factors contributing to the wealth or poverty of nations. 

J. Diamond.  (Apr 25, 2012)  Three Reasons Japan’s Economic Pain is Getting WorseBloomberg View.

Japan is the world’s most distinctive wealthy nation, the non-European nation that industrialized most successfully, and (in its Meiji restoration of the 19th century) perhaps the outstanding modern example of successful selective change at the national level.  But for several decades now, Japan’s economy has been lagging.  Economists and politicians have proposed economic fixes, but these have not born fruit.   I argue that Japan’s economic problems result instead from deeper social problems involving demography, the role of women in Japan, and Japanese attitudes towards immigration and towards management of world resources. 

Genetics, Anthropology, Geography, and Linguistics

J.M. Diamond & P. Bellwood. (2003)  Farmers and their languages: the first expansionsScience 300:597-603.

The Australian archaeologist Peter Bellwood and I had separately written about the spread of the world’s major language families, such as Indo-European and Sino-Tibetan.  As I had discussed in my book Guns, Germs, and Steel, and as Peter had discussed in his own papers, the spread of many language families appears to have been propelled by the origins of farming: the first farmers acquired a big advantage in population numbers, technology, and political organization that enabled them to spread at the expense of hunter/gatherers.  In our 2003 paper, Peter and I collaborated to review the state of this farming hypothesis.  The clearest examples of farming-propelled spreads of language families are the expansions of Austronesian languages from Taiwan across the Pacific, and the spread of Bantu languages from tropical West Africa though sub-equatorial Africa.  Farming-driven language spreads are less clear among Native Americans, whether because of greater uncertainty about New World language relations, greater geographic obstacles to farmer spreads imposed by the New World’s north/south axis, or both factors.  One major question about Old World languages concerns Afro-Asiatic languages, now overwhelmingly concentrated in northern Africa: did they arise in Africa and spread to Southwest Asia against the direction of agriculture’s spread, or did they instead arise in Southwest Asia, spread with farming to northern Africa and diversify there, and then die out in Southwest Asia except for a few Semitic languages (even those possibly being recent arrivals)?  Even more controversial is the most intensively studied, most debated, and seemingly most intractable problem of historical linguistics: did Indo-European languages originate in Southwest Asia and spread with farming, or did they originate in the Ukraine and spread with horse-based pastoralism?

N. Wolfe, C. Panosian, and J. Diamond (2007)  Origins of major human infectious diseasesNature 447: 279-283.

Smallpox, measles, and other epidemic infectious diseases of Old World origins played a major role in the conquests of Native American and Australian and Pacific Island peoples by Europeans, because Europeans but not those native populations had some acquired and genetic resistance to those diseases through prior exposure.  Hence the diseases inflicted far higher mortality on previously unexposed Native populations than on their European would-be conquerors.  Why didn’t native populations harbor equally deadly infectious diseases with which to infect Europeans in return?  In Chapter 11 of my book Guns, Germs, and Steel I discussed evidence that many of those epidemic infectious diseases had evolved from diseases of large domestic animals such as cattles and camels, and that the Old World had 13 species of such domestic animals while the New World had only one and Australia had none.  Hence the evolution of epidemic infectious diseases was geographically one-sided.  In 2007 my colleagues Nathan Wolfe and Claire Panosian and I reviewed newer information about the origins of the world’s 25 infectious diseases with the biggest impact.  As the 1997 evidence had suggested, we found that infectious diseases of the temperate zones (regardless of whether they also infect tropical populations) mainly run in epidemics, because victims quickly either die or else recover and acquire a long-lasting complete immunity against re-infection.  Those diseases are mainly derived from diseases of Old World domestic animals.  In contrast, most major tropical infectious diseases (such as malaria) do not run in epidemics, because they do not induce long-lasting complete immunity and they can thus re-infect recovered victims.  They are also predominantly of Old World origin, but for a reason different from the reason underlying the Old World origin of temperate infectious diseases: they arose from diseases of our closest animal relatives (apes and Old World monkeys), which are confined to the Old World.