The Mounds of Brazil

As I mentioned a couple of posts back, I recently spent a week in Brazil.
I always imagined that when I finally made it to the Southern hemisphere, it would be this amazingly wonderful naturalistic experience. I’d get off the plane and gaze at the stunning peaks from the Lord of the Rings movies. In the near-distance a herd* of kangaroos would bound towards old-growth eucalyptus forest. A bit further away, penguins would be leaping off jagged cliffs into the ocean and just off the tarmac ancient clumps of Welwitschia would be growing in the sandy soil. Maybe I’d have landed just after dusk, and the Southern Cross would be rising in the evening sky.
*Pack? Flock? Gang? What is a group of kangaroos called, anyway?
It wasn’t anything like that.
This was a work-trip, and the company doesn’t pay me to go out and commune with nature. They pay me to sell stuff to big companies in high-rise buildings in cities. Sao Paulo is either the first or second largest city in the Western Hemisphere, and flying in the sheer scale of the place is awe-inspiring- an endless sea of towers riding out of the smog.
Tangent: Only SciFi buffs will get this one. Years and years ago, I read Asimov’s Foundation Trilogy. The capital of the Galactic Empire was the planet Trantor, whose entire surface was covered by one, continuous, planet-wide city. We landed in Sao Paulo just as the sun was rising, and as I looked out of the window, my first thought was “Trantor.”
I spent most of the week in Sao Paulo, where most of the companies we sell to (banking, financial services) are headquartered. The city is remarkable not only for its scale, but for its contrasts. In urban Brazil First and Third Worlds exist side-by-side, and as you go about the city on your business, you cross between the two repeatedly. In First World Brazil, you meet in offices where you’re served coffee in china cups and saucers, and dine in elegant restaurants atop skyscrapers, or ensconced in opulent multi-level shopping malls, watch over by private security staff. In the Third World Brazil you step quickly around piles of trash, watch your pockets and stay off the streets after dark.
Extra Detail: In my office here in Salt Lake City, out of 45 coworkers, I know of only one who’s been mugged*. In my company’s Sao Paolo office, which has about 35 staff, a colleague with whom I spoke could not think of a single coworker who had not been mugged. I could do an entire post full of nothing Brazilian robbery-tales heard over the week, with everything from roadblocks to teams of gun-toting motorcycle bandits who prey upon motorists stuck in traffic.
*It was Matt, who was mugged while on an LDS mission in, yes that’s right, Brazil.
Speaking of motorcycles, a few posts back I mentioned the wide variance in levels of risk tolerance accepted by me and my physician. I experienced a similar disconnect in risk-acceptance in Sao Paulo, but this time with me on the caution end of the spectrum.

Wherever you go in SP or Rio, motorcyclists are weaving in and out of traffic. My favorite moment- which I failed to get on film- was when a motorcyclist as lane-splitting alongside our taxi at ~70 KPH, maybe 12 – 15” from my door, and as he did so was dialing a cell phone with his left hand!

Speaking of taxis, they’re remarkably clean, efficient and courteous in Brazil, as well as very reasonably-priced. They put American taxis to shame. Another area where Brazil puts us to shame is air travel. I took 3 domestic flights in the country. All of them- and the accompanying airport-check-in/security experiences- were delightful- well-organized, on-time and courteous.
Tangent: I should mention that I very much liked the Brazilians I met, worked and interacted with during the week, and that specifically, they were a pleasure to sell to. By “pleasure to sell to…” BTW I don’t mean that they were particularly easy to sell to, but simply that I enjoyed doing business with them. When you’ve sold in a dozen or so countries to at least as many nationalities, certain countries/peoples stand out as more or less pleasurable to do business with. One of my favorite peoples to do business with for example is the Dutch. They’re courteous, straightforward, open-minded and pragmatic. One of my least favorite peoples to do business with is the English*- despite the common language and history I find their indirection and evasiveness exasperating. The same qualities that make both their humor and their literature so wonderful- full of subtlety and irony- make them a royal pain in the ass to hammer out a deal with.
*Before you get on my case for being all anti-Brititic** or what-not, my paternal grandparents were British subjects who met and wed in the UK before emigrating to the US. My father was a dual US/UK citizen until his 40’s. My ancestors include Howe’s second-in-command at Saratoga and a passel of other Englishmen who spent their lives working, fighting and/or dying across the British Empire. My maternal grandfather fought for the British army in WWI. I’ve been to England several times and have generally liked the Brits I’ve known. I just don’t like selling to them.
**I am making up one new word per post.

Nested Tangent: Like many Americans, I’m tempted to award 1^st place to the Canadians, but am ruling them ineligible due to my own bias. As an American I inevitably perceive Canadians as being more or less like Americans, except way, way nicer*.
*Note to Canadian Readers: I’m sorry. I know that annoys the crap out of you when we describe you as “nice Americans”, but it's the truth. That’s really how you seem to us.
And it seemed that the Brazilians enjoyed doing business with me. Wherever I went I was warmly received, and people seemed almost oddly charmed with me. On my fourth day in Brazil I found out why. Ever since growing my beard, I’ve taken it for granted that my doppelganger is Mahmoud Ahmedinejad. But in Brazil it appears I have an even closer double: Rafinha Bastos (pic right), one of the country’s most popular stand-up comics, and co-host of the TV show Custa o Que Custar (“Whatever it Takes”), a weekly comic/ironic news roundup of current events.

Co-workers Matt and Sid claim I even sound a bit like him.
After 4 days in SP we flew up to Rio for the day, and then on to Brasilia, the capital, for a day and a night. Brasilia lies several hundred kilometers North and inland from Rio and SP and is situated in what’s called the cerrado- a type of neotropical savanna, most of which is now farm or rangeland*. Brasilia is a planned city which didn’t exist until the late 1950s. The architecture of the city is striking- a 1950’s vision of what the future would (or ought to) look like. Brasilia feels sort of like someone plopped down the set from a Jetsons episode in the middle of a savannah.
Side Note: Historically, the cerrado has been a pretty crappy place for agriculture, with chronically acidic and nutrient-poor soils. But in recent years Brazil has made huge progress in making the cerrado agriculturally productive, both through new farming techniques and modification/adaptation of alien crops, such as African grasses and Asian soya, to the cerrado climate. A detailed briefing can be found in the August 28 issue of The Economist.
After the crowding, bustle and craziness of SP and Rio, Brasilia seemed wide-open, organized and clean, with wide roads, minimal traffic and clear skies. It was dry- not unlike Utah- and I felt oddly at home there. We spent the day driving between the offices of various government-owned banks, pitching our wares. After lunch we squeezed in a quick tour of the sights for the Americano, including drive-bys of the Palacio do Planalto (President’s office), National Congress, Supreme Court, and various museums and monuments. The tour ran a bit late, and my colleagues urged the taxi driver to hurry to our next appointment. As we did so, we were driving alongside a recently-burned field when I noticed unusual earthen mounds protruding from the ground, mounds that I’d seen countless times in photos, but never before in real life. I’d spent nearly a week in malls, offices, taxis and airplanes, and now, finally, I was seeing something really cool…
“Stop! Tell him to pull over!” I cried. We slowed to a stop and I bounded out of the cab, sprinting across the charred cerrado in my dress shoes, jacket and tie flying. Moments later one of my colleagues, Ricardo, joined me, and together we examined the remarkable termite mounds of the Brazilian cerrado.

All About Termites

What everybody knows about termites is that they’re little white bugs that eat wood, and they live in colonies, like ants. So they’re basically little white ants that eat houses, except that’s not really what they are at all.
All termites eat cellulose, but only a minority of species regularly attack houses. Termites (pic right of worker, not mine) have been around at least since the Cretaceous, and likely since the Jurassic- long before houses were available to eat. They digest cellulose with the help of micro-organisms in their gut, and what’s interesting is that these micro-organisms take on different forms. In about 40% of termite species, the gut-assistants are protozoans, single-celled eukaryotes, meaning that, like us, they have “standard” cell structure with a well-defined nucleus. But the in other 60%, the gut-assistants are prokaryotic (bacteria). These gut assistants can comprise up to 1/3 of the weight of a given termite, and are “transmitted” from individual to individual by trophallaxis, which in the case of termites generally occurs (ew, ick) anus-to-mouth. Termites are eusocial insects, meaning that they are divided into distinct breeding and sterile worker (and sometimes soldier) castes, like ants and (many) bees, and like those insects live and work together in large colonies.
But termites (pic left of soldier, not mine) are different from ants and bees in a number of ways, not least of which is their gender-demographics. In ants, bees and wasps, all workers are female. Males only exist as drones, who basically mate (or try to mate) and then promptly die. Termite workers are both male and female, generally (but not always) sterile. And termite colonies have not just a queen, but a king!
Termite reproductives, or alates- male and female- leave the nest upon maturity to seek out mates, just like the drones and virgin queens of ants, bees and wasps. But unlike those insects, after the termite couple hooks up, the male doesn’t drop dead. Instead the couple goes off together to attempt to found a colony. The king not only continues to live, but also to mate with the queen throughout their long* lives. It’s sort of the Ozzie and Harriet version of the ant/bee/wasp story.
*Up to 17 years in captivity.
On the cerrado termites initially nest underground, but as their colonies grow and expand, they grow not only down, but up. Termite mounds, though new to me, occur over large stretches of South America, Africa and Australia. Though they’ve been known to reach heights of 9 meters, 1 to 3m is more typical. Calling it a “mound” grossly understates the complexity and function of the structure. Termite mounds leverage updrafts to cool the interior of the mound. And even more remarkably, over the course of the day termite workers will selectively open and close specific tunnels to control airflow and regulate internal temperature. This is important not just to manage the temperature of the nurseries, but also- in many cases- of the fungal gardens. Many termite species raise fungi of the genus Termitomyces, which obtain nutrients from the excrement of the termites. The mounds of such fungal-farming species contains carefully temperature-controlled farm/latrine chambers where the fungus grows.
Extra Detail: In Northern Australia the species Amitermes meridionalis builds mounds that are tall, wide and very thin, almost like tombstones. These mounds are always oriented on a North-South axis, and it’s believed that the termites are able to sense the Earth’s magnetic field. The North-South alignment maximizes the exposure of the mounds to the sun at dawn and dusk, but minimizes it at mid-day.
Gazing across virtually any un-developed field around Brasilia you’ll spot dozens and dozens of these mounds, hinting at the millions and millions of termites within and below. In large areas of the tropics termites occupy a similar niche as do earthworms in temperate regions as primary tillers of the soil. And due to the amount of vegetable matter they consume, they’re regarded as pests, not only for crop damage but because they so quickly remove nutrients from the already nutrient-poor tropical soils.
Side Note: One other, much less scientific, but surprising, personal observation for me was just how hard the mounds are. Viewed from a distance they look sort of sandy and crumbly, almost like they ought to wash away in the rain, but to the touch they feel like rough concrete pillars- you’d need a hammer and chisel to take one apart*. According to Ricardo, some years ago the Brazilian government funded a study to determine and possibly replicate the chemical structure of the mounds, with an eye to using the substance as a base for roadbeds.
*Also according to Ricardo, one thing that does routinely take them apart is the Southern Tamandua, Tamandua teradactyla, one of four species of anteaters native to the Americas. Its claws are well-adapted to tearing into termite mounds.
Termites can create such structures (diagram right, not mine) because, like other eusocial insects, they cooperatively work together in large numbers. Eusociality is fascinating not just in its scale, but in its origins. How did it come about? Why would some insects not reproduce so that their mother and (a few of) their siblings might?
This same question actually bugged Darwin, who cited the evolution of eusociality as a special difficulty in his theory of natural selection. Several decades later, a neat explanation was proposed based on the reproductive genetics of ants, bees and wasps, which, together with sawflies, comprise the order Hymenoptera.
Hymenopterans use a haplodiploid system of gender-determination, which means that while females have 2 sets of chromosomes, males have just one. Imagine if instead of X and Y chromosomes, human females had 46 chromosomes (like they do now) but males had just 23. This arrangement leads to some interesting implications for how “family members” are related to one another. Since a male bee or ant has only 1 set of chromosomes, then every one of his sperm cells contains every one of his genes- because that’s all he’s got. A female bee or ant has 2 sets, and so each of her ova contains ½ of her genes.
So a queen bee goes on her (one and only) mating flight and mates with a drone. She receives a bunch of sperm, which she then stores internally for the rest of her life, using it to fertilize (potentially) many thousands of eggs. Her daughters- workers and new queens alike- will receive ½ of her genes and all of their father’s genes, which means that on average they will share- between sisters- ¾ of their genes*.
*But they’ll share only ¼ of their genes with their brothers. This is because haplodiploid males have no fathers; they’re the result of unfertilized eggs. The brothers share ½ their genes with each other, and- oddly- with their sisters, which makes sense when you think about it.
This means that hymenopteran females are more closely-related to their own sisters than they would be to their own offspring- if they could have them- which leads to the so-called haplodiploid hypothesis of eusociality: If you’re a hymenopteran female, the best way to propagate more copies of your own genes is to help your mom make more sisters. It’s a great little story, a triumph of the modern synthesis of Darwinian natural selection and Mendelian genetics*, first(?) proposed by W.D. Hamilton back in the 1960’s, and told countless times since, in textbooks, popular science books, YouTube videos and blogs**.
*Which is why Darwin couldn’t have figured it out, not being exposed to Mendel’s work.
**Including this one, 2 years ago, when I talked about bees.
But the haplodiploid hypothesis doesn’t always hold up quite so tidily in the real world. For starters, many hymenopteran queens, including honeybees, mate with multiple- sometimes dozens- of different males on their mating flight, making their daughters only ¾ or ¼ - related to one another. On the other hand, this might not be an issue if eusociality had evolved prior to polyandrous mating.* But thousands of species of hymenopterans are not eusocial, and, perhaps more troubling, eusociality has evolved in non-hymenopterans, and non-haplodiploid animals, such as… termites.
*And in fact this paper in 2008 claimed to show that all hymenopteran eusocial lines originated from monogamous-mating ancestors.
Termites are diplodiploid, like us; both males and females have 2 sets of chromosomes. And eusociality has arisen in beetles, shrimp and even mammals (naked mole rat, Heterocephalus glaber), all of which are diplodiploid.
In recent years, the consensus on the origins of eusociality has focused more on the broader issue of kin selection (also pioneered by Hamilton), defined as “the evolution of characteristics which favour the survival of close relatives of the affected individual”*. From the perspective of kin selection, the question is whether and when, from a genetic perspective, eusociality is a good deal.
*John Maynard Smith’s definition.
Extra Detail: I’m way understating the current and ongoing controversy around the origins of eusociality. Just 2 weeks ago, 3 Harvard biologists- including E.O. Wilson- published a paper in Nature arguing that natural selection explains eusociality just fine without resort to kin selection. The paper has generated strong criticism from, among others, Richard Dawkins and Jerry Coyne.
Consider an individual’s relatedness to the next generation*. A termite worker is 50%-related to its reproductive brothers and sisters, and only 25%-related to their offspring. For eusociality to be a good deal from the perspective of the worker, its reproductive siblings have to produce way more nieces/nephews than it (the worker) would be likely to produce offspring “going it alone”. (At least 2:1, though that’s an oversimplification, as we’ll see in a moment.)
*By “next”, I don’t just mean successive rounds of siblings, because at some point its queen/mom is going to die. You have to consider its “downlines”**, which since it has no offspring, are its nieces/nephews.
**Sorry. Living in Utah, it is inevitable that a titch of MLM-ese would make its way into the blog.
A hymenopteran worker is 75%-related to its reproductive sisters, and so 37.5%-related to their offspring. Given that it would be 50%-related to its own (hypothetical) offspring, the “good deal” niece vs. daughter ratio for a worker bee would be considerably less than 2:1, or more like 1.33 (50/37.5). So while haplodiploidy may not be a requirement for eusociality, would it make it either likelier to occur or easier to sustain? (The ratio is unrealistic because it ignores that in a non-eusocial context, her sisters- potentially all of them- would still be producing nieces/nephews as well, just not as many of them.)
I don’t know, and it is, as I’ve mentioned, a way controversial topic in evolutionary biology, but a possible hint may lie in the occurrence of eusociality. At present it’s thought to have evolved independently at least 17 or 18 times*. (New instances have been discovered just in the last couple of decades, and I imagine it’s likely more will be…)
*By “times”, I don’t mean species. There are probably a million(?) species of animals in the world with eyes, but eyes are believed to have evolved independently maybe several dozen times (~40-70). Similarly there are thousands of eusocial animal species, but the evidence is that eusociality itself has evolved only around ~17 or 18 times.
Of those 17 or 18, 11 were within Hymenoptera. Another 1 was with Thrips*, who although not closely-related to Hymenoptera, are also haplodiploid. Another instance was with aphids, who are cyclically parthenogenic, leading to inter-relatedness issue at least as weird and compelling as those in haplodiploidy. So 13 out of 17 (or 18) instances of eusociality appear to be linked with haplodiploidy or parthenogenesis. Kin selection may not depend on haplodiploidy, but it doesn’t seem to hurt.
*Order Thysanoptera. Teeny-weeny insects about 1mm long that root around in the soil and eat (mostly) stuff like fungal spores. 2 cool things about them. First, they’ve been around since the Permian; they’re one of the kinds of creatures that survived the cataclysmic Permian extinction, which made the K/T extinction event (dinosaurs) look like a rainy day at the park. Second, they have just one mandible, always on the left.
Of the remaining instances, 1 or 2 occurred in mammals (Naked Mole Rat, and possibly the Damaraland Mole Rat, Fukomys damarensis). 1 occurred in crustaceans (sponge-dwelling pistol shrimp, Synalpheus regalis), 1 occurred in beetles (Austroplatypus incompertus) and 1 in… cockroaches.

Roaches?!

Which leads us to the weirdest-thing-I didn’t know-about-termites: they’re cockroaches. Not just “closely-related to…” cockroaches, but cockroaches. Here’s a cockroach family tree. Cockroaches only form a monophyletic group if you include termites.
Most Americans, if they’re familiar with any cockroaches, know the German Cockroach*, Blattella germanica, and the American Cockroach**, Periplaneta americana. An American Cockroach is more closely-related to a termite than it is to a German Cockroach. And both the German and American Cockroaches are more closely-related to termites than either is to the desert cockroaches.

*Which infested all of my college dorms. Sometimes late at night, I wonder which destroyed more brain cells during my college years- maryjane or the chemical residue from those roach-killer fog-bombs I used to detonate in my room periodically.
**Which infested the house outside of San Diego where I was born. Despite the name, it’s native to the Old World. In does well in the tropics and has been spread by shipping, which explains why it does so well in a warm-climate port like San Diego.
Side Note: See the Mantids branch of the family tree? Yup, that’s right- a Praying Mantis is pretty much a big green carnivorous roach.
The cockroaches most closely-related to termites are the Wood Cockroaches (genus = Cryptocercus). Wood Cockroaches live and care for young socially (but not eusocially) and digest cellulose with the aid of internal gut protozoans, which are closely-related to those inside of termites, and which the roaches transmit anus-to-mouth to their young. This required transmission has been suggested as a possible factor in roach/termite social evolution; without extended parent-offspring care/association, such transmission might not occur. Termites are wood roaches that became monogamous and took the next step to sterile worker and soldier castes. It’s a very different path to eusociality than that followed by the hymenopterans, who appear to have evolved from social predatory wasps.
I’ve talked a lot about mounds and termites without mentioning the species of the mounds I checked out. It’s not easy to know; the termites only come outside/above ground at night, but more importantly, there are well over a dozen common mound-inhabiting species in the Brazilian cerrado. Researchers in the 90’s found that over a third of termite mounds in the cerrado contain more than one species of termite; some housed up to 14 different species! The relationships between these cohabitants are not entirely clear, but in at least some cases appear to be mutualistic. Different species of termites have different defensive mechanisms. Cortartermes sp., for example, a common mound-cohabiting species, has soldiers which specialize in chemical warfare. Perhaps multiple modes of defense compliment one another, benefiting the various cohabiting species.
Many mounds also contain 1 or more ant species. In some cases these are abandoned mounds, but in many others they’re mounds currently occupied by 1 or more species of termite as well. In some of these cases it appears that the ants may be playing a mutualistic defensive role in the mound, which is curious in that ants are generally the most reliable enemies of termites (the Tamandua notwithstanding.) In any case, there’s a lot happening inside those mounds.
From the taxi Marcio yelled at us to hurry; we were late for our appointment. Ricardo and I turned away from the termite mounds of the cerrado and jogged back toward the human mound of Brasilia.
Note about sources: An especially helpful paper was Evolution of Eusociality in Termites, Barbara Thorne. Additional information on eusocialty came from Eusociality: Origin and consequences, Edward O. Wilson and Bert Holldobler. Phylogenetic info on cockroaches/termites came from Death of an order: a comprehensive molecular phylogenetic study confirms that termites are eusocial cockroaches, Daegan Edward et al. Info on polyspecific mounds came from Ant and termite mound coinhabitants in the wetlands of Santo Antonio da Patrulha, Rio Grande do Sul, Brazil, E. Diehl et al. Info on ancestral monogamy of eusocial hymenopterans came from Ancestral Monogamy Shows Kin Selection Is Key to the Evolution of Eusociality, William O.H. Hughes et al. Additional info on- and specific examples of-eusociality came from this online course module developed by Zachary Huang of Michigan State University. The termite caste graphics used in the family tree were pulled from a pest control company website- I forget which one.