Thursday, June 30, 2011

Ned Kock Provides a Philosophical Breakthrough on Fruit

by Chris Masterjohn

In retrospect, I don't know why this wasn't obvious to me years ago, but my hindsight is really the only thing that didn't start failing me when I turned 17 and my opthamologist told me I was "just getting older."

Ned Kock recently provided me with a philosophical breakthrough on fruit, buried in a post about boring food.

In 2007, I wrote the following in my Thyroid Toxins Special Report:
While anti-milk writers are quick to point out that cow’s milk is not meant to be a human food, one rarely encounters a writer who points out that leafy vegetables – that is, leaves – are not meant to be foods for any animal, but are meant instead to carry out photosynthesis for the survival of the plant from whom they have sprouted. Indeed, fruit and milk are the only foods that the producing organism “intends” to be food for animals; of the two, only milk is meant to nourish rather than attract the animal who consumes it. Since plants are immobile, they produce a variety of defenses that mobile animals do not need: some of them are structural, such as thorns; others are chemical and include a wide variety of toxins. Humans have developed ways to detoxify many such plants when used as staples, to selectively breed less toxic varieties, or to utilize some of these toxic components for specific medicinal purposes.
This reasoning always bothered me a little bit.  Why, then, are fruits nutritious?  Why do so many more people report problems with milk than report problems with fruit?

Granted, there is a lot of biological variability in humans, and a lot of new food intolerances thrown on top of that variation because of our industrial diets.  Some people don't do well on fruit, perhaps because they consume it with fat (though I love berries in my yogurt), don't absorb fructose well, or have insulin resistance that needs to be corrected.  

Nevertheless, fruit seems to be much more universal than milk.  Heck, even the Inuit ate fruit:
The foods of these Eskimos in their native state includes caribou, ground nuts which are gathered by mice and stored in caches, kelp which is gathered in season and stored for winter use, berries including cranberries which are preserved by freezing, blossoms of flowers preserved in seal oil, sorrel grass preserved in seal oil, and quantities of frozen fish.  Another important food factor consists of the organs of the large animals of the sea, including certain layers of the skin of one of the species of whale, which has been found to be very high in vitamin C. . . . Their fruits were gathered in the summer and stored in fat or frozen for winter use.  (Weston Price, Nutrition and Physical Degeneration, pp. 71, 259).
Ned Kock has solved the puzzle:
Plants want animals to eat their fruits so that they can disperse the plants’ seeds. So they must be somewhat alluring to animals. Sugar plays a role here, but it certainly is not the only factor. The chemical composition of fruits is quite complex, and they usually contain a number of health-promoting substances, such as vitamins. For example, most fruits contain vitamin C, which happens to be a powerful antioxidant, and also has the ability to reversibly bind to proteins at the sites where sugar-induced glycation would occur. . . .  
Two things must be kept in mind regarding fruits and their evolution. One is that dead animals do not eat fruit, and thus cannot disperse seeds. Sick animals would probably not be good candidates for fruit dispersion either. So the co-evolution of fruits and animals must have led fruits to incorporate many health-promoting attributes. The other is that seed dispersion success is correlated with the number of different animals that consume fruits from a plant. In other words, plants do not want all of their fruits to be eaten by one single animal, which must have led fruits to incorporate satiety-promoting attributes.
Plants need to make fruits relatively nutritious, satiating, and non-poisonous, in addition to deliciously sweet  and visually alluring, because it is the only way to get a variety of animals to pick up their seeds and live long enough to poop them out somewhere else.

Of course, none of this philosophizing can tell us what the effects of fruit are on health or longevity, or how much if any fruit we should consume.  

Plenty more on that in the future.

But next up, a series on the Masai, including sexuality and gender roles, their knowledge and use of hundreds of plants, whether they had any heart disease and how sure we can be, and why they had such massively low cholesterol, almost as low as Pygmies and Bushmen.  

Also coming up soon, can normalizing thyroid hormone abolish the risk of heart disease?  What's with that temperature test that everyone either loves or hates?  Plus a series on measuring and interpreting blood lipids, a couple of posts addressing whether insulin can block leptin signaling and whether blood triglycerides could be a starvation signal, why fat can cure carb cravings and when and if it may sometimes be better to just eat carbs, and lots more.

And none of that is meant to neglect the nowhere-near-done The New Genetics series (which will get much more interesting when I get into the controversial stuff) or the upcoming My Genetics series.

Stay tuned!

Read more about the author, Chris Masterjohn, PhD, here.


  1. Ummm Ray Peat basically wrote the same thing quite a few years ago

  2. Hi Anonymous,

    I did say it seemed rather obvious so I'm sure Ned's not the first to ever formulate the idea, and I did fault myself for missing it. Could you post a link to where Ray Peat said this?


  3. Why satiating? Also, I would expect toxins or something associated with the seeds to prevent or discourage mastication or digestion and to increase motility.
    Fruit color is a big clue too. Trying to attract attention is a way to be eaten...

  4. Hi Ed,

    His argument, which I believe I quoted, is that it increases the chance of seed dispersal success to have multiple types of animals and multiple animals eating them. Thus, if they are somewhat satiating, it increases the number and probably type of animals eating them. I agree with you about the seeds, but most fruit seeds are small enough to pass digestion and miss mastication, so thus become irrelevant. When I wrote "visually alluring," I was referring to their color.


  5. Honey is also intended to be food. Just Saying.

  6. When I originally read the post on Ned's blog, I wondered why this didn't also apply to grain seed. My assumption -- after running thru a number of examples -- is that grain seed doesn't survive our digestion (or preparation methods) intact. Not sure if that universally applies tho.

  7. Actually, there is a whole science of seed dispersal, which distinguishes between seed dispersers and seed parasites. A seed parasite chews or eats the seed, which is why seeds are bitter to discourage it, or spits it out after eating around it, foiling the plant's effort to get the seed moved to a new location. The "intended" seed disperser should be the right size animal to swallow the seed whole and defecate it elsewhere. This has been used to show that some plants are "evolutionary anachronisms," where extinct megafauna (usually wiped out by humans) have originally spread the plant. These plants have become restricted in range or propagated by humans. A good example is an avocado, whose seed we eat around and don't accidentally swallow, like the extinct giant ground sloths would. "Substitute" megafauna have been suggested to reinhabit regions, such as Paul Martin's idea of reintroducing elephants and camels to replace the still missing ecological role of seed dispersers in the American West, which would diversify and restore the plant ecology. See Connie Barlow, The Ghosts of Evolution: Nonsensical Fruit, Missing Partners, and Other Ecological Anachronisms (Basic Books, 2000).

  8. Most animals create their own vitamin C, I believe. Humans and guinea pigs are two that do not, but I doubt most fruits know that.

    Not picking on you or Ned, just pointing that out.

  9. Thanks Chris, and I look forward to the series on the Masai.

    You are correct Jennifer. Fruits seems to have evolved to "work" for the animals that could disperse the plants' seeds. Possibly many species with different nutrient needs. Vitamin C is only one of them.

  10. So one question that comes to mind is the following: If fructose is far worse for us than glucose, than why do plants provide sweetness with sucrose (a mix of glucose and fructose), rather than pure glucose?

  11. Don't seed-hulls have inhibitors that prevent them from being digested? Isn't that why the WAPF recommends soaking and sprouting all seeds?

    So this makes sense, with the fruit being relatively-easily digested, and the seed itself being protected as it passes through.

  12. Hi Beth, as others have said, I think that in general seeds are designed *not* to be eaten, while fruit flesh is designed to be eaten -- the end result that the seeds remain undigested while the animal carries the seed off somewhere else.

    Bill, thanks that was really interesting.

    Jennifer, I'm aware of that, but I don't really think it affects the point since there are many nutrients in fruit and Ned only used vitamin C as an example, and one that is of course very relevant to us humans.

    Ned, you're welcome, and thanks.

    Albert, I don't know quite enough about plant physiology to answer your question definitively and I'd hate to reduce a plants entire existence down to feeding animals as if it didn't have other reasons for synthesizing compounds, but frankly I don't buy into the "fructose is far worse than glucose." I think the relative benefits or harms of fructose are highly contextual.

    Leigh, yes I think you've got it.


  13. Oh and anonymous, good point. I suppose I had overlooked honey because it has such an ambiguous status as a regurgitated plant food derived from animal labor. It's difficult to decide whether to give primacy to the activity of the bees or the plants that created the initial nectar.



  15. One quick comment about fructose that is in line with Chris's above. In animals with livers depleted of glycogen, it turns the liver into a "sponge", greatly increasing the rate of glycogenesis based on both fructose and glucose:

    If your liver is full of glycogen, fructose is not going to be so good. But if your liver is full of glycogen, you should not normally feel hungry for those natural foods that contain fructose.

  16. It's not just the muscles and the liver that store glycogen, but the brain does as well.

  17. Anonymous,

    Thanks, I had read that article prior to writing my Thyroid Toxins report. At the time, I was reading some of Peat's writing and some of Katherine Milton's writing, and they were both influences on me to a certain degree when developing that idea about fruit. However, I don't think Ray Peat really formulates the idea as fully as Ned has. He does say in the article that they are nutritious, but he only connects the lack of toxins back to the adaptive advantage of the plant. He doesn't explore the benefits of good health (as opposed to lack of death), number of animals and variety of animals in producing a nutritious and satiating product the way Ned did so succinctly and explicitly. I don't mean to suggest that Ned is the first person to ever have thought of this -- I doubt that is the case -- but he's the first person who clearly explained it in a way that made that light bulb go off in my head as just now.


  18. I'm the anonymous who mentioned honey.

    It seems to me that fructose only occurs when plants are trying to attract animals (fruit, pollen). When plants make carbohydrates for their own ends, they use glucose (starch).

    Maybe fructose evolved as a lure for animals.

  19. "blood triglycerides could be a starvation signal"

    when i was diagnosed with anorexia, my trigs were real low

  20. Anonymous, maybe. Malibu, "could be" is the critical phrase there. ;-)


  21. I've always thought capsicums were a cool example of this... or the opposite of this, actually. If what I recall is correct, their is a theory that hot peppers produce capsaicin as a way to discourage humans (and other mammals) from eating (and digesting) their seeds. The digestive system of birds, however, does not destroy pepper seeds, who happen to be immune to capsaicin and happily eat the hot peppers and spread the undigested seeds in little piles of fertilizer... Or so goes the theory!

  22. Interesting ideas.

    I believe there are only two substances in nature intended to be nutritious for animals (insects excluded): milk and egg yolk. The egg yolk is the only food that developing birds, reptiles, etc. have when they are in the egg shell, so it better be nutritious. Similar story for milk. Every other substance seems only nutritious by coincidence, including fruit.

    I fully agree on the points of fruit having to be satiating, and non-poisonous. And yes, it is useful if fruit provides some calories to help fuel the physical act of seed dispersal. But nutrition beyond the calories appears to be coincidental in my view.

    Why does fruit have antioxidants? The word already gave it away: oxidation protection. Fruit that oxidizes easily, i.e., fruit that rots quickly, is less likely to be eaten → seed dispersal chances ↓ → evolutionary disadvantage compared to fruit that keeps well for a longer time.
    Which selective pressure is likely to be highest and most consistent? Increased seed dispersal chances by meeting the widely varying (micro) nutrient needs of different species of animals that eat the fruit? Or increased seed dispersal chances by the fruit staying well over a longer time? I opt for the latter. Usually evolution is quite egoistic.

    Now, given that this is Cholesterol-and-Health: just as lipoproteins come packed with antioxidants to extend the time that they can safely circulate in the blood, fruit comes packed with antioxidants to extend the time that it is alluring enough to be eaten.

    Why are most of the antioxidants in/near the peel? The peel is where the fruit is exposed to oxygen, UV, and large temperature swings, i.e., factors that increase oxidation. The inside is exposed to very little oxygen or UV. And the temperature variations inside the fruit will be dampened, by for example the water content. That means less antioxidants are required within the fruit.

    I guess that if food manufacturers would be forced to only use naturally occurring substances, they too would use things like vitamin C, polyphenols, phytochemicals, etc. to extend food's shelf life. And to make efficient use of the substances, they would likely apply them liberally on the outside, and use the bare minimum for the "core".

    Long story short: I think that fruit is nature's candy which comes packed with "selfish" preservatives. Some of these preservatives can be very useful for the human body, and so it makes sense to eat fruit (if tolerated).


  23. Wow John, I swear you carbophobes/fructophobes will do anything, and I mean anything, including creating some science fiction theories inside your brain to justify silly ideas like fruits are just natures candy or insulin spikes cause diabetes. At first it was kinda funny but now it's just getting sad...

  24. John,

    I agree with you that it would be overly simplistic and almost certainly wrong to say that everything in a fruit is made for the animal eating it, but I think you are going a bit overboard by anthropomorphizing "egoism" into evolution. A fruit does not have an "ego." It cannot be "selfish." If an effect on an animal's health results in a more successful dispersion of seed, the plant doesn't have some "selfish" opposition to the process. The seed grows if it lands in the right conditions, and the plant doesn't have any choice. Its hereditary material is perpetuated not because the plant wanted it to be, but because the seed grew.


  25. While Ned's arguments certainly stand pre-agriculture, it is no longer the case that fruits that are more nutritious/satiating etc will be selected.

    As fruits are now cultivated, they are (in the US and UK at least) selected by growers largely based upon their appearance, shelf life, and resistance to pests.

    Don't get me wrong, I'm not anti fruit by any means, but it does bear thinking about.

  26. "Anthropomorphize" is now my new favorite word :D. And of course, great article. Fruits are certainly a food that some paleo peeps say should be limited or sometimes even eliminated, an assertion I find ridiculous. They certainly need a better reputation as does everything actually produced and palatable without the assistance on modern agriculture and processing.

    With that said, John's argument certainly made some since as well. If natural selection means survival of the fittest, it would make sense to me that trees that randomly evolved to have longer lasting fruit would be more likely to survive than ones with more nutritious fruit granted both tree's fruit were palatable. This is probably over simplifying things, but I'm sure this is what he meant by egoism; the trees propensity to survive based on the laws of natural selection. In all reality, it is probably a combination of both factors, evolution has been occurring for a long long time.


  27. Jeff,

    I think you hit the button here: "In all reality, it is probably a combination of both factors."


  28. Simon,

    Have you read Denise's post on this subject?


  29. Chris, Thanks for the link love buddy. Great post as always.

  30. "Two things must be kept in mind regarding fruits and their evolution. One is that dead animals do not eat fruit, and thus cannot disperse seeds. Sick animals would probably not be good candidates for fruit dispersion either."

    OK, I think its pretty obvious its not a good idea for plants to poison the animals that disperse their seeds. At least not acutely. Aside from anything else, its a waste of the plants resources. Why produce expensive poisons when you can just give them some cheap sugar?

    "So the co-evolution of fruits and animals must have led fruits to incorporate many health-promoting attributes."

    Wow hold your horses there Ned. Thats a pretty huge jump you've made there, from saying plants shouldn't poison their animal seed-dispersers, to concluding that they therefore MUST make them healthier. I'm not saying its impossible that some plants might have evolved attributes to promote the health of the animals eating their fruit, but to say it "must" have happened seems like a wild exaggeration.

    The most obvious explanation for plants to pack their fruit with costly antioxidants, as John has explained, is to prolong their "shelf life" and so increase the change of being eaten and having their seeds dispersed.

    The hypothesis that an additional or alternative selective force acting on most fruit-bearing plants was to make their animal consumers healthier is hard for me to believe.

    Imagine a population of fruit-bearing plants, where the fruits are essentially just "cheap bags of sugar" around the seed to attract animals to eat them. Then imagine a mutation arises in one plant that results in it packing extra nutrients into its fruit, to the health benefit of any animal eating it. However, assuming that these nutrients are "costly" for the plant to produce, it will be expected to suffer a cost, perhaps by producing less (but more-nutritious) fruit.

    A monkey (or whatever) comes along and eats the fruit. It gets a big dose of vitamin C (or whatever) and so he's going to live a long and healthy life, free of scurvy. He then goes over the next tree (which doesn't have the mutation, and so produces more, but less-nutritious, fruit) and eats the fruit there. And then to the next tree, and so on. Other monkeys and animals do likewise.

    At the end of the fruit season, all the monkeys are nice and healthy and free of scurvy thanks to our mutant tree, and they have done a good job dispersing seeds. But what thanks does the tree get? It produced less fruit than all the other trees (who also benefit from all the healthy monkeys to disperse their seeds too), dispersed less seeds, and so ultimately the mutant gene is selected AGAINST in this population.

    "The other is that seed dispersion success is correlated with the number of different animals that consume fruits from a plant."

    You got any evidence to back this up?

    "In other words, plants do not want all of their fruits to be eaten by one single animal, which must have led fruits to incorporate satiety-promoting attributes."

    Since when have fruits ever been satiating?

    And if they are, what are these "satiety-promoting attributes"? Extra sugar? Fat? Calories? Sounds expensive. If I was a plant, all I'd care about was getting my seeds dispersed as cheaply as possible. And if thats by one single fat little greedy monkey, that's fine by me. Sure, other things being equal maybe I'd slightly prefer them to be eaten by a number of different monkeys, but I don't really care enough to spend any extra energy making "satiety-producing attributes" to make it so.

  31. Basically what I'm saying is that if a mutation arose in a plant that made it spend extra resources to provide more nutritious fruit for animals, all the other trees in the population would out-compete it by free-riding on the benefits provided by the mutant (i.e. healthier animals*) without paying any of the cost.

    Unless the mutant could some how keep all the healthy animals to itself and not "share" them with the other trees. Even if the monkeys could somehow detect that the fruit from this particular tree was more nutritious, once it is out of fruit they would just move on to the next.

    * To be honest, even this seems kind of unlikely. A monkey eats a fruit, then moves on and shits out the seed within a day. Is it gonna die of scurvy or any other disease in that time (that could have been prevented by extra nutrients in that particular fruit)?

    I doubt it!

    I just don't see the incentive for the plant to spend valuable resources to "take care" of the monkey. Just spend the bare minimum (some cheap sugar) to attract a hungry animal to come eat your fruit, then job done.

    Unless there is a big over supply of fruit and not enough animals to eat them, and each plant has to maximise its attractiveness to the choosy monkeys. But I can't see that happening too often in nature. There will always some hungry animal out there grateful for any scrap of food, even a fruit that is no more than a cheap bag of sugar.

  32. Hmmm, what I find "getting sad", and not even "kinda funny":
    1) Anonymous wrongfully placing me in the carbophobes/fructophobes category. Fact is that I don't do low carb, and I eat fruit almost daily. I think the idea that insulin spikes cause diabetes is silly, just as I find fruit avoidance due to fructose silly.
    2) Anonymous using terms like "you carbophobes/fructophobes", as if learning more about health were a matter of "us" vs. "them".

    The point of my comment was that I believe fruit can be nutritious and useful. I just doubt that the main purpose of the nutritional value beyond the caloric content (e.g., antioxidants etc.) is for the benefit of animals eating the fruit. It would mean that the plant would have genes encoding for complex enymatic pathways to synthesize substances, which only serve the plant itself in an extremely indirect way. Also, in order for an animal's nutritional needs to be a significant selection criterion for the plant, these nutritional needs seem to have to be fairly consistent for a long period. This appears unlikely to me, especially if several different species eat from the same plant.

    I find it more plausible that most of the beneficial substances in fruit are there so the fruit stays well for a longer period. This increases its chances of getting its seeds dispersed. And these substances just happen to benefit animals that eat the fruit, i.e., they are nutritious by coincidence.

    But I'm perfectly willing to change my mind on this purely philosophical and speculative idea. What would it take? Examples of fruit substances which do not appear to serve the purpose of fueling seed dispersal (calories), increasing palatability, increasing visual allure, or decreasing potential oxidative damage. In other words, substances that appear to only serve the purpose of nourishing the animal that eats it (which in the end would of course serve the seed dispersal too).


    That is indeed what I meant with egoism. I forgot to put quotes around it like I did with the word "selfish".
    I agree with Chris about fruit not having an "ego". The words "egoistic" and "selfish" just seemed like convenient ways to describe what I meant, but that didn't quite work out ;-)


  33. Anonymous,

    I agree that the idea that a plant *needed* to put vitamin C in something is preposterous, but I took Ned's use of "must" in a very different context -- one pondering, "hmm, I wonder what happened? Oh, this must have happened." I think that is a quite different use from "This must happen or the plant will die" or "You must now go do _____."

    Your scenario involves an awful lot of assumption, and you ask for evidence without having any. Who said fruits were initially bags of sugar? If scurvy is severe, it can cause death. Micronutrients may play a role in food intake and satiety, so it's quite possible that selection could act through an animal's choice and appetite (mediated by instinct) to.

    I think what we are outlining here is possible and sensible selective pressures.


    I think the evidence you are asking for on this "purely philosophical and speculative idea" is kind of high. I agree with you that preserving the fruit tissue would be selected for. The problem is so many substances could do both.


  34. John and Anonymous,

    To put this another way, let's turn it around and look at it from the perspective of the fruit-eater. There are many species that base their diet on fruit. I gave some examples in my "primitive wisdom" post, like the agouti, paca, and acouchi. A number of primates, as I'm sure you know, are largely frugivores.

    No one is debating why nutrients initially found their way into the fruit. That is interesting, but the question is whether the health of frugivores affects the nutrition of fruits through selective pressure.

    So the question, then, is this: are these plants better off with or without these frugivores? If fruits were not nutritious, these frugivores could not base their diet on them. Presumably, they have their own sets of instincts that lead them to select a somewhat nutritious and sustainable diet. Moreover, they can select different fruits, causing competition among the plants.

    The argument here is that the fruit-bearing plants are better of with the frugivores than without. Thus, since the frugivore's success depends on their health, the health of frugivores mediates selective pressure to eat a somewhat nutritious and sustainable diet. This in turn mediates selective pressure on the plants to make them appealing to the frugivore.

    The alternative argument, that the plant only puts sugar into the fruit for the frugivore but vitamins into it for itself, assumes a great deal of efficacy for natural selection on the plant, and a great failure for natural selection on the animal eating it.


  35. Chris,

    I like your idea of looking at it from the perspective of the frugivores. In my view there are a couple of assumptions in it:
    1) Most frugivores that have access to the fruits have instincts to guide them to proper nutrition.
    2) There are different sorts of fruit to choose from.
    3) The different fruits have different nutritional profiles.
    4) Most frugivores are guided to the same types of fruit (similar nutritional profile), creating competition between the fruits.

    Certainly I agree that if these assumptions are true, this can mediate a significant and consistent selective pressure on the plants to make them appealing to the frugivores.

    A list of my assumptions (of course I'm biased, so probably I'm "conveniently forgetting" some):
    1) There are frugivores that have access to the fruit.
    2) The fruit that has more oxidation protection lasts longer.
    3) The fruit that lasts longer has a higher chance of getting eaten.

    Somehow I still feel that these assumption are true more easily than the ones my biased mind came up with based on your description. That certainly doesn't "prove" I'm right. It might just show that I'm stubborn and repetitive :-)

    Luckily our speculations don't have to be mutually exclusive. Also, I admit that if fruit has more oxidation protection, it is likely to be more nutritious. In that case the question becomes: was the fruit selected because it lasted longer, or because the frugivores were instinctively guided to them, or a little bit of both? We can only speculate.

    Now, I also have a scenario: one species of frugivores and one type of fruit.
    Who will adapt to whom? My guess is that the speed of mutation is the main factor: if the frugivores mutate fastest, they will likely adapt to the fruit; if the fruit mutates fastest, it will likely adapt to the frugivores.


  36. Chris,

    "I agree that the idea that a plant *needed* to put vitamin C in something is preposterous, but I took Ned's use of "must" in a very different context -- one pondering, "hmm, I wonder what happened? Oh, this must have happened." I think that is a quite different use from "This must happen or the plant will die" or "You must now go do _____."

    I also interpreted Ned's sentence this way, but I just disagree with the idea that plants producing antioxidants etc in fruit "must have happened" because it benefits the health of the frugivores, as if there is no other possible explanation. I believe there are other possible explanations (prolonging fruit "shelf life") and that these are to me more plausible.

    As I mentioned before, I am not ruling out the possibility that improving the health of frugivores could have been an additional or alternative selective force for some plants, I'm just disagreeing with Ned's (and your??) assertion that it "must have happened". That is all.

    "Your scenario involves an awful lot of assumption"

    The only assumption I made was that it was costly to the plants to produce these various nutrients/antioxidants, i.e. costly in terms of energy, or nutrients, or evolving enzymatic pathways, etc. That is a fairly reasonable assumption I think. But maybe you disagree?

    "and you ask for evidence without having any."

    I was asking if Ned had any evidence to support his specific claim that "seed dispersion success is correlated with the number of different animals that consume fruits from a plant."

    I didn't make any specific claims in my post. All I did was take issue with the assertion that plants must have evolved antioxidants etc in their fruit to improve the frugivores health, then ask him to think about the actual process of evolution and plants 'competing' against each other in terms of the number of genes each provides to the next generation, to explain why I think his hypothesis has some difficulties.

    "Who said fruits were initially bags of sugar?"

    I think it is parsimonious to assume that when fruits first evolved, they were more likely to have been "bags of sugar" (with any antioxidants there to protect the fruit from oxidative damage) rather than full-blown nutritional power houses designed to improve the health (beyond providing cheap calories) of any animals that decided to eat it. It just seems easier for me to believe that a mutation could arise that resulted in a tree covering its seeds in a "bag of sugar", rather than believe that the original mutation was instead "cover the seeds in a bag of sugar and also produce lots of nutrients that will improve the health of any animals in the area".

    But perhaps I am wrong. Perhaps the simple mutation (bag of sugar) did arise first, but the hungry animals who came across this early fruit took a sniff of it and thought "nah, there's no nutrients in that, its just empty calories, I'm not going to eat that. I'll just walk around for another few hours and hope I come across something more nutritious".

    If that did happen, then yes the simple "bag of suagr" mutation wouldn't have spread, and we would have had to have waited until the "sugar and loads of nutrients" super-mutation came along before we ever saw animals willing to eat them, and fruits finally taking off.

    "If scurvy is severe, it can cause death."

    I would imagine the chance of a monkey dying of scurvy in between eating the fruit from a particular tree, and shitting the seed out a day or so later, is so small that is would a pretty much negligible selective pressure on a tree to provide vitamin C in its fruit for this reason.

  37. "Micronutrients may play a role in food intake and satiety, so it's quite possible that selection could act through an animal's choice and appetite (mediated by instinct) to."

    Yes its "quite possible" I agree, but "must have happened"? I'm not so sure...

    If the environment was saturated with an overabundance of fruit-producing trees, and there was not enough animals to eat them all, then yes animals could maybe afford to be choosy about which fruit they eat, and there would be selective pressure on the trees to be more attractive to the animals.

    If the animals could reliably assess the vitamin C (etc) content of different fruits and chose their fruit accordingly, then plants could be selected to increase their vitamin C content. That's a lot of assumptions though.

    Also, this is a simple "attract the frugivore" trait, which is different from Ned's hypothesis that plants produced these nutrients to ensure the local frugivores were healthy and so in good shape to disperse their seeds - which is a much more indirect selective pressure. The problem with the latter is that the healthy monkeys will disperse everybody's seeds, and the free-riders will likely win out over the tree that actually produces the most-nutritious fruit.

    Ironically, the hypothetical situation in which the selective force will be strongest for the plant to improve the frugivore's health is when a tree has a very close relationship with a single species (or better yet, a single indivdual) who is it's exclusive seed disperser, and who in turn eats only from this one tree (i.e. they are completely mutually dependent), as opposed to when the fruits are eaten by various species, themselves probably generalists who obtain their food from a wide variety of sources.

    Yet Ned seems to suggest that plants do the exact opposite of this by making their fruit "satiating" and so ensuring no one individual or species of animal eats all its fruit, thus leaving more for the generalists to come in and munch.

    Anyway, it is a fascinating discussion and I'm glad you brought it up Chris. I enjoy reading your blog (if health blogs were premiership football teams, yours would definitely be one of the "big 4" along with Melissa's, Stephan's and Matt Stone's). Keep up the good work!

  38. OK let me use an analogy that may be more familiar to many people reading this blog. Instead of why do plants produce antioxidants/micronutrients in their fruits, why do industrial food manufacturers add these things to their products?

    In this case, the initial "bag of sugar" fruit model, is replaced by some processed food made of some combination of grain, soy, vegetable oil and sugar - basically the cheapest product that is tasty enough to attract a consumer to want to buy it.

    There are 3 hypotheses that we have discussed so far in relation to the fruit, which are not mutually exclusive. Here are their industrial food analogies:

    1. Preserve shelf life - eg adding ascorbic acid (vitamin c) to processed food.

    2. Some customers value the nutrient content of their food, are trying to eat healthy etc, and will actively choose products with more nutrients than other competing products that are not so endowed. So the food manufacturer adds extra vitamins, antioxidants, omega 3, whole-grain fibre, or other things customers want, or makes other modifications (reduced saturated fat/cholesterol) that will increase their attractiveness to customers. These things may or may not actually make the customer healthier - the food manufactuer doesn't care, as long as there is demand for it (selective pressure), it will provide it.

    3. Ned's hypothesis: The food manufacturer adds extra nutrients etc because it genuinely wants to make the population healthier. It figures that if it makes American people healthier, and less people die of scurvy, diabetes, heart disease, cancer, etc, then there will be more people alive to buy its products and so maximise its profits. Therefore the manufacturer is happy to spend a bit more money to make its products more nutritious (and take a hit on its short-term profits) to achieve this, as in the very very long run it will increase sales and boost profits.

  39. Ned's other hypothesis: Food manufacturers will make their food as satiating as possible so that any one single consumer won't buy too many, leaving more on the shelf for other customers to buy later.

  40. ...because sales success is correlated with the number of different customers that buy food from a shop

  41. Even the Inuit ate fruit .... perhaps, but it can't have been a very large part of their diet in general, and, except at certain latitudes, it can't have been an important part of most early humans' diets 365 days of the year.

    Some posters have said that they eat fruit every day, and indeed, modern man can do this, but this does not mean that mankind evolved to eat fruit 365 days of the year.

    It may indeed have been vitally important for the fruit that some (any) animal ate it and passed the seeds in its waste products, but that does not mean that it was ever vital for any individual species of animal that it should eat fruit.

    As Chris says "Of course, none of this philosophizing can tell us what the effects of fruit are on health or longevity, or how much if any fruit we should consume."

    I speak as one who used to eat a lot of fruit (to the point of addiction), and became overweight and pre-diabetic. I'm not going to take a chance on fruit again. Of course this doesn't happen to everyone.

  42. Don't really know how to explain this, but anyone think that maybe these frugivores can actually taste nutrients and therefor seek/eat the fruits from the trees highest in them, spreading seeds from those? If the tree produces less nutritious fruits, less of it's fruits are eaten. Over hundreds or thousands of years that less nutritious strain of tree would be phasing out pollinating with the more nutritious ones.

    I have no doubt frugivores who've been basing their diets based around fruit [x] would be highly sensitive to which trees have the most nutrients they need. They know the good stuff, just like a beer snob or wine connoisseur.

    Humans have terrible instincts, but even raw foodists can be sensitive to the taste of particular nutrients they may need. For example, a need for sodium will have them seeking fruits/veggies highest in that like tomatoes.

    The plant doesn't know what the animal needs, but the animal does. VERY SLOW selective breeding going on here.

  43. Chris, an obliquely related question: I recall some of your posts on native nutrition about treating digestive problems, involving experiments with SCD-type regimens, coconut oil fasts, etc. Now I have been following with interest your recent arguments that foods containing starch, fructose, gluten, etc., can potentially be health-promoting in traditional, whole-food forms. So here's my question: if you had to treat a digestive problem (like IBS etc) now, what approach would you take?

  44. John makes a lot of sense.
    Comming back to the comment about fruit dispersers and fruit parasites by Bill: doesn't this mean we as humans will be best adapted (or the fruits have not developed special toxins towards us) to fruits where we are indeed fruit dispersers, like grapes, tomatoes(?), pommegranate, watermelon (maybe the occasional cherry stone), in general fruits with seeds so small to be easy to swallow for us and survive our digestion?
    And if we are fruit parasites for something like peaches, plums, avocados, apples (we don't eat the core usually and if we do the soft seeds are easily chewed and destroyed) why those plants wouldn't protect themselves from us?
    What about berries? Do the seeds survive our mastication in all/most/a few cases?


  45. Responses to John, Anonymous, Montmorency, Grok, Max, and Franco.

    Hi John,

    It seems the assumption you have the most difficulty with is that frugivores have some instinct guiding them to a nutritious product. I'd like to restate this in a way that might be easier to accept: frugivores would not be able to survive or reproduce on a diet that was not nutritious; therefore, frugivores that instinctually eat a diet that is not nutritious will be selected out; only those that eat a sufficiently diet will remain. The fact that some species instinctually eat largely fruit-based or exclusively fruit-based diets is a rather definitive proof that selective pressure either to produce or to maintain fruit in a nutritious state exists. I agree that our theories are not mutually exclusive, and for this reason I can only accept "a little bit of both" as an answer to the question you pose.

    I don't really understand your scenario. If there is one species of frugivore and one species of fruit, hypothetically, it would have to be true that the fruit is nutritious for the precise reason that the species of frugivore would not otherwise be able to survive and reproduce.


    I don't believe it is parsimonious to believe that fruits were anything when they "first evolved" unless there is some type of evidence of that. It could seem equally simple for a 'fruit' simply to be a covering of existing tissue that already has nutrients in it for use of the tissue. A tissue can't metabolize anything without nutrients, so it would be a greater deviation from "existing tissue" to produce a "bag of sugar."

    I think it is a rather narrow view to consider selection taking place on a rather immediate basis in a single event between eating a fruit and leaving the seed behind. If animals that cannot synthesize vitamin C develop instincts to eat a diet that does not contain sufficient vitamin C, they will be selected against by dying before the produce a maximum number of offspring. You are injecting all kinds of unneeded assumptions into this scenario like the animal's ability to "reliably assess the vitamin C content of different fruits." The animal simply dies from scurvy, or can't reproduce effectively. Period. Selection on the frugivore does not have any temporal relationship to when the frugivore eats the seed. It occurs at any time between birth and the reaching of maximum reproduction. Instinctual selection of diet then mediates selection on the plant separately.

    I realize you are trying to introduce an analogy for the purposes of communicating this simply, but I think the choice actually obscures what is going on by differentiating between what is intentional and unintentional. The problem here is that when speaking of selection, all of it is intentional.

    In any case, thank you for your comments. I'm glad you're enjoying the blog and you find this discussion interesting.

  46. Montmorency,

    I didn't say that the Inuit ate a fruit-dominated diet. I simply used them as an example of a group that consumed fruit even where plant foods are very limited in the environment , as an anecdote demonstrating the fact that fruit consumption is more universal than milk consumption.

    Regarding whether early man could have eaten lots of fruit, please see Denise's post, which I linked to above.

    Regarding whether "it was ever vital for any individual species of animal that it should eat fruit," it is of course vital to the frugivores that do not eat anything else. I listed three examples above.

    As you said in a comment on another post of mine, "To be fair, I had also in my life consumed a vast amount of dried fruit e.g. dates, figs, raisins, etc, and I suspect that might have been what did the most damage." I think this was likely the root of your problem with fruit.


    I find it plausible that some species may have a taste for nutrition, but I don't think it's necessary to assume this in order to make the argument that food selection would mediate selective pressure on the plant for nutrient value. Regardless of the precise mediator of food selection, animals that do not select nutritious food will be selected against.

    Max Renn,

    I think gluten is difficult to digest and I'm not sure gluten-containing foods could ever be central to restoring digestive health (though I suppose it could be possible), but I have spoken with people who normalized digestion by getting away from anti-starch dogma in GAPS and SCD and including corn and potatoes along with eating to satiety, and the improvement seemed to be downstream to normalizing thyroid status. I do think restricting type or quantity of carbs can be a useful tool, but not a cure-all, when disacharide digestion is compromised.


    Since I don't think we exclusively feed on many fruits, apart from other species, I don't think we should assume we'd be better adapted to the fruits that we mediate selective pressure upon. The main way we do that now in any case is artificial selection.


  47. My comment to Anonymous above should say "all of it is UNintentional." Sorry about that.


  48. This comment has been removed by the author.

  49. Hi Chris,

    To be honest, I don't have a difficulty with the assumption that frugivores have some instinct guiding them to a nutritious product. In fact, I find it fairly implausible that they don't. I just note that the assumption goes hand in hand with the assumption that they have a choice. Natural selection doesn't seem to work that well if there are no alternatives to select from ;-)

    Certainly it would not be optimal if frugivores don't have a choice and the fruit available to them doesn't meet their nutritional requirements fully. But I believe there are many shades of grey between dying out due to malnutrition and thriving by being fully nourished. A frugivore whose nutritional needs are met for the most part but not fully, probably will have offspring. In turn, this offspring might have genetic mutations which make it better adapted to the nutritional profile of the fruit that is available.

    Which brings me to the scenario I described (not too clearly). I speculate that mutation speed and reproductive cycles will influence "who" "exerts" most selective pressure.
    Let's take berries as an example, and let's assume that the berries allow the frugivores to "just survive" (nutrition wise it's not optimal, but sustainable). If these berries are eaten and their seeds are dispersed in the first year, with a bit of luck there will be new berry plants in the next year. These new plants may have mututations that could change the nutritional profile of the berries a bit. This means that even just one generation of frugivores gets exposed to many genetic permutations of fruit, i.e., there is substantial choice and thus the frugivores can select.
    In this example, due to fast plant mutations, it is likely that over time the frugivores select for more nutritious fruit, allowing them to go from "just surviving" to "thriving". So the plants adapt to the frugivores.

    Now, if we replace the berry plants by some fruit bearing trees, the cycle becomes a lot longer. It can take years before a dispersed seed becomes a tree which can provide a reasonable amount of fruit. So it will take longer for slightly mutated fruit to become available. This seemingly reduces the amount of selection the frugivores can do (at least initially). If the period between dispersal and fruit bearing becomes very long, it seems more likely that frugivore mutations appear which can better handle the available fruit. When that happens, the frugivores adapt to the trees.

    In other words: the species that mutates fastest, seems most likely to become adapted to the species that mutates slower. You could say that the species that mutates fastest, is the most flexible species.


  50. There's no satiaty promoting effect of fruits. It's in the best interest of the plant to induce hunger on the consumer. If the consumer consumes more of the fruits, the seed dispersal probability rises, it's the same effect for industrial food which are laced with sugar. IMHO it's a secondary effect of fructose. It has been shown that fructose in not to big quantities promotes better insulin sensitivity, which will make blood glucose drop faster, which will lead to higher appetite.
    Furthermore, we seem to lack an accounting feature for fructose, meaning that these calories are nearly invisible to our systems, making it quite easy to overeat.
    In a natural environment, the (relative) scarcity (or better the seasonality) of fruits and available fructose will not lead to problems, on the contrary it's even a advantage to be able to gorge on big quantities of fructose (and the associated glucose) for a short time. The liver will transform it in its long term storage form: fat.
    My conclusion: eat fruits (or sugary stuff), in season, intermittantly. Don't eat that stuff continually.

  51. Gallier2,

    There has to be satiety value to fruits. If there was not, than the species adapted to exclusive feeding on fruits would eat until they got so bloated their stomach would burst and they'd die. This is not desirable for the plant.

    Here's an example of humans doing this intentionally with milk:

    This does not happen naturally with any foods because they all have satiety value.


  52. Yes, that's true, that there's some satiety value in fruit, but my point was the same as Anonymous made before (and I swear I've read his comment about it after I made the comment): there is no selection pressure on fruit plants to make fruit satiating, on the contrary. There is selective pressure on the fruit eater, as you have rightly observed, to not overindulge, be it only a mechanical reason of stomach capacity. There's also another selective pressure for a frugivore to not overconsume, if they did overindulge and get bloated on fruits, they would be easier preys for predators.

  53. Gallier2,

    You might be right, but I think you're being somewhat dogmatic about it. It is plausible that if a larger number of animals and a more diverse set of species feed on a plant, it will have more success. To demonstrate or refute this, you'd have to do some empirical observation to see whether that is actually a determinant of success in the real world.


  54. I'm not dogmatic, sorry. I'm maybe a bit too assertive in the idea I express, but rest assured that I'm open to change my ideas if presented with good arguments, and you usually provide them.
    The thing is, that I've always been a lousy writer, even in the 2 languages I grew up with (French and German). It was quite easy to learn English, as it is not much more than a germanic dialect with a lot of french vocabulary, but this doesn't helped to solve my initial lack of proper writing skill.

  55. Hi Gallier2,

    Sorry about confusing your tone. I agree we need more evidence on this issue to really demonstrate anything, and right now we are just speculating. Thanks for your contributions!


  56. Chris:

    Many thanks for your reply. Re: Denise Minger's article: Very interesting and enlightening.

    Her point about "Dangerous Natural Substances" does tend to argue, at least a little, against Ned Kock's thesis about fruits "wanting" to be eaten. This did not only apply to unripe fruit, if I read the article correctly. (Would be perfectly understandable for a fruit not to "want" to be eaten until ready).

    Her analysis necessarily is based upon the wild fruits that still exist today. Given that she is talking about our evolutionary journey from Africa more than 50,000 years ago, it would be fair to comment that wild fruits then were not necessarily the same as wild fruits now, i.e. they have also evolved. Maybe it wouldn't make much of a difference to the argument, but perhaps it would have been good for her to comment on this aspect.

    I realise that her main point is (I think) that it doesn't take human intervention to produce a fruit that is sweet and delicious and high in fructose.

    And something that I personally have no knowledge of, but on which experts presumably have a view would be: did we all in fact evolve from Africans who lived "closer to the equator"? Or were we more widely spread out, with not all of us possibly having access to fruit all the year round. And if we started migrating 50,000 years ago, is this a long enough period for us to have evolved away from the people who were used to year round fruit? (And why did we migrate away from this apparently lush landscape - presumably because of climate changes of one sort or another).

    On my personal fruit experience: Yes, it was probably dried fruit that first started to clobber my metabolism, but although I wasn't keeping records at the time, I seem to remember that I had pretty much given up dried fruit for at least a year or so before I realised I had a serious problem that needed to be addressed and at this time, I was still eating a lot (more than "normal") of fresh fruit as well as a mixed, probably high-carb, diet, although no obvious "junk" (no HFCS-sweetened drinks, no added sucrose, not many "sweets").

    My point being that stopping dried fruit didn't stop the problem getting worse (maybe it slowed it down), and I was at least still semi-addicted to fresh fruit at this time.

    Going low-carb, then very low-carb at least ameliorated the symptoms, although I don't claim any kind of cure. If I eat fruit now, even in modest amounts, I put on weight, so I know it can't be good for my metabolism, and I also know that if I start by eating modest amounts, I'll soon increase those amounts, so for me at least, fruit has to be a no-no.

    This metabolism-scarred ugly old veteran has a slightly more jaundiced view of fruit than a pretty, slim 23-year old :-) (I was slim at 23, although never pretty :-) ).

  57. @gallier2, I'm living proof that fruit is satiating. It's literally almost the only thing that I can be satiated from.

  58. Funny, it has the exact opposit effect on me. Except for bananas, fruit gives me the impression of opening a hole in my stomach.

  59. Montmorency,

    I don't think the idea that fruit "wants" to be eaten is Ned's -- I think virtually everyone agrees that fruit flesh and peel largely serves to attract animals that will aid in seed dispersal. Most plant foods contain lots of toxic substances, but these are quite often believed to have beneficial properties through hormesis, the effects will vary between species, they may vary with ripeness, and as Denise notes they may inhibit consumption. If the latter two points are generally true, this provides a mechanism to postpone consumption of the fruit until the time is ideal, just as the seeds within them have mechanisms for postponing germination until the soil conditions and climate are appropriate. If minor concentrations of these compounds remain in ripe fruit they could also provide a mechanism for Ned's point about inhibiting excessive intake.

    I agree that the "out of Africa" theory is likely to be simplistic, especially given the latest genetic analysis that is much more consistent with an "into Africa" theory. My impression is that this will lead to a hybrid with the archaeological evidence into a "humans went out of Africa and came back in" theory, but then if you're willing to accept questionable claims about supposedly human teeth, you could put a different spin on the archaeological evidence.

    Here's an enlightening article on the paucity of evidence for a clear understanding of human origins

    "Did early Homo migrate “out of ” or “in to” Africa?"

    He offers several scenarios but notes "it would be misleading to claim that
    any of the scenarios are supported by that meager evidence."

    I also agree with you that we cannot assume these fruits are truly wild in the sense that humans have exerted no selective pressure on them. We have to remember that 1) horitculturalists are not agriculturalists but they also are not hunter-gatherers and they do in fact exert selective pressure on plants and that 2) hunter-gatherers also exert selective pressure on plants by gathering them.

    I think the main value in Denise's article is in showing that sweet and delicious fruit is not a product modernity. Whether it existed in the early paleolithic seems irrelevant. If this selection occurred over that time period, it means humans co-evolved with the fruit. People forget that what differentiates us from other animals, dietarily, is mainly the need for huge amounts of calories to supply our big brains. Our big brains are what allow us to breed foods with a higher ratio of calories to toxic, fibrous, and irritable components so that we can fuel the development of those same big brains.

    We should realize that while excess calories are an issue nowadays, the epidemic of obesity and type two diabetes is decades old, not thousands of years old, and our high calorie-demanding brains predate these epidemics by eons.

    I agree with you that people will vary in their response to fruit, but I don't think it's just whether someone is young with a healthy metabolism or old with a damaged metabolism. Some people claim to reverse metabolic problems by going low-carb or zero-carb but others do so by going fruitarian. Any of these approaches might be flawed, even seriously flawed, but they all seem to have some merit at least for *some* people and at least in the short-term. There is no grounds for claiming that anyone with metabolic damage needs to restrict carbs or abolish fruit. I do agree that your experience suggests you should restrict fruit, and I'm glad you've achieved a healthier state by doing that.


  60. Grok and Gallier2,

    Great exchange. And thus we see that everyone's a little different. :)


  61. So, fruits are nutritious because they "want" to be eaten, while seeds contain toxins to discourage their consumption.

    I presume that plants also do not "want" to be eaten. Why is it that vegetables seem among the most healthy things for us to eat?


  62. Albert,

    Because they are rich in nutrients, and also sometimes toxins we are well adapted to that have beneficial properties through hormesis. Things that have compounds that inhibit appetite, of course, could also be good for you by discouraging excessive consumption -- clearly it is a complex topic.


  63. Are seeds also not rich in nutrients? Or is it the excessive consumption of the seeds (which is made possible by modern day processing), that tips the balance from good to bad?

  64. Seeds also contain nutrients. I didn't say anything about seeds being bad. However, seeds are often rich in PUFAs, so I think it is best to use them in moderation. They are also rich in anti-nutrients such as phytate, so it is probably best to process them in some way like soaking or sprouting, particularly if your diet is very marginal in certain minerals such as calcium, iron, zinc, etc. If used in moderation on a nutritious background diet this is probably not much of a concern.


  65. Thanks for the reply. I was mostly thinking of the negative things I've read about seed oils when formulating my questions. Do you think the problems with the oil come mostly because of the PUFAs or do the anti-nutrients also play a large part?

  66. In the oils, PUFAs, see here:

  67. Chris,

    Thanks for your reply.

    Quote: There is no grounds for claiming that anyone with metabolic damage needs to restrict carbs

    Really? Not even Type 2 Diabetics? Now that does surprise me.

    (I'm not talking about my personal issues here).

  68. Hi Montmorency,

    I think my choice of grammar was rather poor. I'd have communicated my point better if I said "there is no grounds for claiming that *everyone* with metabolic damage needs to restrict carbs." In other words, some people may help reverse their metabolic damage by restricting carbs, but in other cases this might not work, or something else might work better.


  69. If all food is satiating, then non satiating stuff is *not* food.

  70. Fruits are really more universal and also more nutritious. By eating fruits regularly, you can really feel that your body functioning will get better and better. These are very powerful substances.

  71. @Chris,

    Not sure if you will necessarily see this as it is now a fairly old article.

    I hadn't read anything by Ray Peat when I made the postings above (despite his name being in the 1st comment!), but discovered him a short while ago. A fascinating, and of course controversial gentleman (having heard him speak, "gentleman" seems the appropriate word to use).

    For what it's worth, partly influenced by him, I've been experimenting very cautiously in re-introducing some fruits, so far with no deleterious effect. I won't be going near dried fruit or commercial fruit juices, but hand-squeezed and strained orange juice (in necessarily small quantities) doesn't seem to have turned me into a juice addict. Leaving aside for the moment whether the addition of the sugars is beneficial on an otherwise VLC diet, I think the vitamin C, potassium and smaller quantities of other minerals can probably only be good.

    Peat is very selective about what fruit he considers acceptable, and given what most of us actually have available, it doesn't leave much. One of his points is that fruits that might otherwise be fine are not fine in their current "industrialised" state, e.g. the residues from chemical treatments of one sort or another causing allergies or other problems.
    (e.g. commercially-grown apples, and bananas - the latter are also starchy unless overripe. He's not keen on starch).

    I'll keep reading and studying and cautiously experimenting, and try to keep an open mind, but my views on fruit are much more "nuanced" than in recent years.

    We probably should be careful not to regard all fruit as being equally good (or bad), just as we should not regard all carbs as being equally bad (or good).


To create a better user experience for everyone, comments are now moderated. Please allow up to one business day for your comment to post. In order to avoid the appearance of spam, please avoid posting links, especially to commercial destinations, and using all-caps.