Thursday, September 3, 2009

Primal Potatoes, part 2

In addition to our high level of amylase production, we also have another feature that raises questions about idea that Paleolithic people ate a strictly carnivorous or starch-free diet: we store glycogen, a starch, in our liver and muscles. The typical human can store 100 g of starch in the liver, and 300 g in muscle tissue. This implies that natural selection favored the survival of people who could store starch in amounts up to 400 g at a time¬––an amount impossible to get from non-starchy vegetables. What purpose could this serve?

Supporting this feature, we have the insulin system. Although many paleodiet thinkers focus on the lipogenic effects of our insulin response to carbohydrate, lipogenesis via insulin only occurs if we have no immediate use for the glucose and we have full glycogen stores. If we don’t have full glycogen stores, insulin will convert glucose to glycogen first. This interaction between insulin, dietary glucose, and glycogen indicates that human physiology evolved in an environment in which conversion of dietary glucose into glycogen stores conferred a survival advantage.

Starch, Glycogen, and Hunting Success

Glycogen storage appears to have a significant effect on physical performance. Phinney investigated the effect of a ketogenic diet on physical performance, and found that, given adequate adaptation time and mineral nutriture, a ketogenic diet can support endurance performance equivalent to a high-starch diet. However, he also noted that athletes on ketogenic diets show decrements in anaerobic performance—e.g. weight training, weight lifting, or sprint performance, particularly at the end of an endurance event. This occurs because sprinting or lifting requires the production of ATP in the absence of oxygen, and human mitochondria cannot derive ATP from fat in the absence of oxygen. Thus, we require intramuscular storage of glycogen to retain and develop maximum strength or sprint performance, as needed in the “fight or flight” response.

Aerobic or steady state endurance activity has very little impact on glycogen stores unless carried out for one hour or more. In contrast, high intensity activity--like sprinting or resistance training--rapidly depletes muscle glycogen. Whereas typical glycogen stores will support an intense aerobic exercise for a few hours, a single maximal sprint effort will deplete one-third to one-half of glycogen stores (Fournier et al, 2004).

Humans can replenish glycogen stores without dietary carbohydrate, and even while fasting (Fournier et al, 2004), which emphasizes the importance of glycogen stores for survival (specifically, in fight or flight situations). In the absence of food, lactate and endogenous amino acids provide the substrate for glycogen replenishment--you sacrifice lean mass for glycogen. In the absence of dietary carbohydrate, dietary amino acids would supply the substrate for glycogen replenishment.

Since we can store approximately 400 g of glycogen, and it takes one gram of protein to produce one gram of glucose/glycogen, in the absence of dietary carbohydrate, replenishment of just one-third of muscle glycogen (100 g) depleted by one maximal effort each day would require intake of 100 g of protein above dietary requirements, or the breakdown of up to 100 g of lean mass––nearly one-quarter of a pound of muscle.

People with physical activity levels similar to hunters doing persistence hunting require 1.0 to 1.6 g of protein per kilogram of body mass to maintain nitrogen balance if eating a mixed diet containing adequate glucose (Tarnopolsky, 2004). A 150 pound (68 kg) individual would thus require 68 to 108 g of protein to maintain lean mass assuming a mixed diet containing carbohydrate. If his diet had no carbohydrate, he would require 133 g additional protein each day on which he made efforts requiring him to replenish just one-third of his total (muscle and liver) glycogen. So, his total requirement would be at least 201 up to 241 g of protein daily to prevent loss of lean mass. This would require the consumption of about 28 to 34 ounces of lean meat – 1.75 to 2 pounds. If he depleted one-half of his glycogen daily, he would require 268 to 308 g of dietary protein, or 38 to 44 ounces of lean meat (2.4 to 2.75 pounds) every day.

Such an individual would expend about 3000 calories daily. An intake of 201 grams of protein provides 1064 calories, 27% of the total caloric intake. An intake of 241 g of protein would provide 32% of calories. This gets near the maximum intake of protein possible without exceeding the liver’s ability to detoxify the ammonia that results from deaminating the amino acids in the process of converting them to glucose. An intake of 308 g of dietary protein would provide 1232 calories, 41% of total caloric intake, an amount that would likely cause protein poisoning.

These apparent facts raise the intriguing possibility that the use of tubers fueled human success in hunting. Simply put, people who had fuller intramuscular glycogen stores would have superior sprinting, wrestling, and lifting ability compared to people without them. Such ability certainly would not improve human ability to harvest tubers, since they don’t run away or resist capture. But I imagine that when humans hunted on foot, with spears and such, using some type of persistence hunting, those who had the best “kick” into a sprint would frequently have had greater hunting and reproductive success than those who did not.

Consider this feat of strength recounted in The Paleolithic Prescription:
“In 1805, the Lewis and Clark expedition witnessed an Indian bison kill ….A small herd was stampeded over a cliff into a deep, broad ravine. As the bison fell one on top of the other, dazed and injured, hunters killed those on top with spears; the others were crushed and suffocated underneath. The ravine was twelve feet wide and eight feet deep; most of the bulls weighed over a ton, yet a team of five Indian hunters pulled nearly all the bison out of the ravine onto level ground for butchering.”

This is all anaerobic activity. Our current knowledge indicates that men who had more stored glycogen would have had greater success in feats like this, compared to men who did not.

In addition, if attacked by a predator, natural selection would have favored the survival of those who had a strong “kick” fueled by glycogen over those who had less stored glycogen and limited sprint ability.

I can think of several other advantages people using starch from tubers instead of protein from meat to supply glucose for replenishing glycogen would have realized:
1. Lower dietary protein/meat requirement, reducing the pressure for success in hunting large animals, and making it possible to feed more people (offspring) with each kill.
2. Less burden on the liver for ammonia detoxification.
3. Easier to avoid protein poisoning while at the same time maintaining greater glycogen stores.
4. Easier to maintain and increase lean mass in response to the stresses of high intensity activity, with a lower dietary protein requirement.
5. Reduced pressure to hunt only the fattest animals by use of carbohydrate instead of fat to dilute the protein content of the diet; which greatly enlarges the pool of potential prey, increasing dramatically the amount of energy available for harvest.

Thus it seems likely that people who ate both meat and tubers regularly and had increased amylase and had the ability to store glucose as intramuscular glycogen, rather than as fat, would have scored more often in hunting, more often avoided predators, had access to more prey, and had ability to support more offspring, compared to those who did not eat tubers, or did not have enough amylase, or did not store glucose from tubers as glycogen.

Glucose, glycogen, insulin resistance, and intermittent feeding

Modern people easily consume 300 to 400 grams of glucose daily, which means that they will always have full glycogen stores unless they do something to deplete them on a daily basis. When muscle have full glycogen stores, they exhibit insulin resistance, but if you deplete the muscles of glycogen, they become insulin sensitive.

I have already noted that hunters engaged in the type of high intensity activity required to deplete glycogen stores and maintain insulin sensitivity. In addition, hunter-gatherers typically ate only once or twice daily. When you fast, the liver store of glycogen gets used up in 8 to 10 hours, and muscle glycogen reduces by about 50 percent in 24 hours. Thus, the primal combination of high intensity exercise and intermittent feeding would likely have maintained insulin sensitivity even with regular intake of glucose-rich starches.

Also consider that Holt et al (see Mendosa’s report) performed a study to determine the satiety value of various foods. White potatoes turned out to have the highest satiety index of any food tested, twice as satisfying as cheese or eggs, nearly twice as satisfying as beef, and about 50% more satisfying than ling fish, measured two hours after feeding. If that to which we have adapted gives us the greatest satisfaction, then this supports the idea that foods like the potato played a very important role in satisfying our ancestors. Since meat also had a high satiety value (nearly 75% greater than that of white bread), roasted meat and potatoes would have provided our ancestors with very satisfying fare.

Thus, I have come to accept that tubers probably played important roles in human evolution, and propose that the use of tubers actually increased success in hunting by improving physical performance. This seems to me the best explanation of all the unique features of human physiology. I think it may explain why many people find meals of meat and potatoes deeply satisfying “comfort” food.

The next question is, will a diet high in tubers adversely affect dental and therefore general health? I will address that question in Primal Potatoes, part 3.


Anonymous said...

Great article Don, thanks. There seems to be quite a debate in the paleo community about starches, and this seems like a winning argument. I personally feel more energetic since adding white rice and potatoes to my otherwise "paleo" diet.

Stephan Guyenet said...


Great post. I like your point about starch opening up new ecological niches. I was thinking about the native North Americans on the East coast who ate corn and lean game. Without the corn, they wouldn't have been able to exploit the game in any quantity because it's too lean for most of the year.

I eat a lot of potatoes. I agree with commenter Aaron in the last post, they seem like a "clean" fuel. Rapidly absorbed, low anti-nutrients, some fiber but not excessive, plenty of vitamin C. Plus the protein quality is quite high, so you don't have to complement it with other protein sources to make good use of it. It also contains a surprising amount of protein (roughly 10% of calories). I also like that they're cheap and totally unprocessed.

One concern I have about potatoes is the glycoalkaloid content. OK, every plant has some kind of toxin, but the thing that's worrisome about these glyoalkaloids is they can increase intestinal permeability under some circumstances. That's one of the reasons why Cordain doesn't include potatoes in his version of the paleo diet.

He provides evidence that they increase permeability of intestinal cell cultures, as well as in a genetically susceptible strain of rats, at concentrations that you would actually get from eating potatoes. I don't know whether it's a concern for humans or not. I feel it's pretty hypothetical at this point. I haven't succeeded in finding good data on the health of a non-industrial culture that relies on potatoes as a staple. Do you have any opinions on this?

However, I do know that in Peru, where the potato originates, they peel them. It turns out, most of the glycoalkaloids are in the skin. There are often reasons behind the food habits of non-industrial people...

Charles R. said...

I, too, had more strength when I added some starches (primarily rice) to an otherwise VLC diet. I was surprised, as I thought it would affect the aerobic more than anaerobic, but your explanation for why that isn't the case makes sense.

I think the rest of your argument seems logical as well. I'll be interested to read the dental discussion, as I have not seen any evidence previously that would indicate starches as causing dental issues, at least in the context of an otherwise low-carbohydrate, balanced-mineral diet.

Dan said...

Very nice post (and timely, I was just talking about this topic with a friend last night).

I have a question/point about the depletion of liver glycogen after 8-10 hours of fasting. I am assuming that this would be used mostly to fuel the brain (assuming a sedentary individual)?

If this is the case, let us assume a person who has followed a very low carb diet for awhile - sufficient time for their brain to adapt to running on much smaller amount of glycogen (40g vs 120g per day). In this case, assuming constant muscle glycogen, they would need to produce a much smaller amount of glycogen daily - perhaps some amount that could be accomplished with protein intake alone.

Of course muscle glycogen wouldn't remain constant unless you were sedentary. You mention that muscle glycogen stores deplete by 50% after a 24hr fast. Does it continue to deplete beyond that, or does it then remain constant? If it remains constant, then even a 50% level would be enough to maintain a max intensity effort for ~10 minutes (say catch a meal or evade a beastie).

Of course nothing that I am saying is going against your original argument - that the ingestion of starch would have conferred an evolutionary advantage.

If indeed this is all true, then a possible way to reconcile starch with a standard "paleo meat-and-fat only" diet would be this: eat starch in preparation (sufficiently long enough beforehand to digest to glycogen) for intense exercise. This allows you to take advantage of the extra glycogen for better workouts. Afterward, allow yourself to remain in a depleted state (ie don't try to pack the muscles or liver full of glycogen afterward - no sports drinks... but hey aren't those half fructose anyway?! :) ). You still get all of the benefits of the low-carb version of the paleo diet, but you also get to use the high-octane rocket fuel when you need it.

And now after re-reading what I wrote, I basically summarized "carb loading". :) Perhaps that isn't such a bad thing if 1) we don't start out with full glycogen stores, and 2) we don't do it chronically and return to being dominantly a sugar burner, and 3) allow the muscle to stay in a depleted state afterward - maybe I am wrong here but I seem to recall some benefits on growth hormone/testosterone by allowing muscles to remain glycogen depleted post-exercise? perhaps I am totally off base on this one

This is pretty much what I do right now, but my choice of rocket fuel is raw whole milk rather than potatoes.

Wow this one got out of hand rapidly.

Chris said...

very helpful post Don - thanks

Sven said...


your argumention implies that carnivorous animals doesn´t use glycogen. As far as I know at least dogs and cats do have glycogen stores.
I think we can rule out the possibility that (wild living) dogs eat a meaningful amount of tubers (or grains). A feat of domestication? What about lions, etc.?

Mark said...

Great post!

So in regards to an optimal diet, I have come around recently that the right carbs (fruit but not too much because of fructose, veggies but not too much because I just don't see them as beneficial to the body, potatoes, corn that has been treated with lime, white rice, and even grains and legumes if they have been treated and soaked properly) should be fine for the body.
I am interested in the protein amount being toxic though. As so many in the bodybuilding community start with 1 gram of protein per pound of bodyweight, many go higher, do you have an idea of what is required for muscle growth/maintenance? I only ask because protein seems to be the only independent factor here. Once the amount of protein is set (at least a maximum), both carbs and fats fall in line how they will.

After looking at the work of Lyle McDonald in his Rapid Fat Loss Handbook, it reinforced how the only essential foods are protein and fatty acids. So if someone is trying to avoid metabolic syndrome while looking good at the beach, I would theorize that they set a calorie deficit while eating good foods. Within this setup, I would set an "optimal" range for protein and fill in the rest of the calories with carbs and fats. Do you think I'm far off?

I work as a trainer in my spare time so I'm trying to find the perfect balance of nutrition between optimal health while looking good. I believe that this merging point will provide both. I think too many people, me included, get wrapped up in the carb-phobia/'Good Calories Bad Calories' theory. Your blog and Stephan's Whole Health Source have helped me tremendously. As a Crossfitter, it's easy to just listen and not question sometimes.

Looking forward to part 3! Sorry for the long comment, I just LOVE this stuff.

Don said...


My argument DOES NOT imply that carnivorous animals don't use glycogen. They generate it from amino acids, just as humans can. They differ from us in their tolerance for ammonia generated in gluconeogenesis. I don't know how much glycogen a cat or dog can store, but I do know that humans can store 400 g, and doing so on a carbohydrate-free diet would present apparently insurmountable problems. So then the question is, what selective pressure would lead to the emergence of humans who do have the ability to store 400 g?

A strictly carnivorous diet would exert a selective pressure for elimination of amylase, a smaller capacity for glycogen storage or a greater tolerance for high protein intake. We see neither in any human group so this argues against the idea that human evolution was driven by a low carbohydrate, carnivorous diet.

We certainly can tolerate a low carbohydrate diet (witness the Eskimos), my argument is that in any environment where starch is available, people who can use it have multiple selective advantages over those who can't, and it is a fact that most habitats occupied by human ancestors did supply starch (tubers). It doesn't apply to dogs or cats, since they don't have a primate background that gives the capacities we have.

Keep in mind that Neanderthals, known to be strictly carnivorous, died out. H.sapiens sapiens didn't. If pure carnivory is more adaptive (in most habitats), why aren't the Neanderthals still kicking?

Anonymous said...

so how do you imply then, from a diet stand point... people who eat a high fat, mod carb and protein diet and gain copious amounts of weight. if the body does not "want" to be fat as i think everyones body wants to be healthy, why are so many people gaining weight with tubers who eat primal, and why are so many people fat?

i see many primal eatters who dont overdo the protein to any extent, and steer clear of any "starch(tuber)" because of the resulting weight gain.

Don said...


I have not seen any good evidence for protein intake greater than 1.4g/kg/d given adequate energy and carbohydate (to prevent the use of amino acids for gluconeogenesis). Tarnopolsky demonstrated that this level maximizes whole body protein synthesis in resistance training individuals.

That translates to 0.6 g/lb, about 95 g per day for a 150 pound individual. You'd find that in about 14 ounces of lean meat. Given a caloric intake of 2500 daily, that amounts to about 15% of caloric intake. So the other 85% can come from CHO and fat. 50% fat and 35% CHO would deliver about 220 g of CHO.

Perhaps I will post an example in the near future, as I have created a diet outline the mimics the typical Aborigine or African hunter-gatherer use of animal and plant foods.

Don said...


"Does it continue to deplete beyond that, or does it then remain constant?"

It depletes, especially if the individual engages in anaerobic activity, and (absent food intake) the body will replenish by using amino acids taken from breakdown of lean tissue.

"If it remains constant, then even a 50% level would be enough to maintain a max intensity effort for ~10 minutes (say catch a meal or evade a beastie)."

And then you have to replenish, either by breaking down your own lean tissue to yield amino acids, or by eating an extra 1 g protein for each gram of glycogen repleted, or by eating carbohydrate.

Don said...

BTW, if we use Tarnopolsky's 1.4 g/kg/d as the requirement for a hunter gatherer, this will strengthen my argument, because the hunter will now need 30 more grams daily of protein just to maintain whole body protein synthesis. This will increase the amount of protein needed to maintain glycogen (on a zero carb diet) by another 30 grams daily -- another 4 ounces of lean meat.

Don said...


Thanks for the comments. I agree with peeling potatoes. I don't have at my fingertips any data on health of white potato eaters. That will fit into my next post.

Mark said...

So in essence, the optimal diet (assuming good food as previously discussed) is one that hits the minimal amount of protein (0.8 g/kg/d to 1.4 g/kg/d) and then the remainder of calories are divided between carbs and fat. Carbs shouldn't be too low (<50g) for too long unless you raise protein to compensate, even then that is not optimal. If you want to get more detailed, more carbs will be needed as you workout (resistance training, sprints, intervals, etc) because your glycogen stores will start to be depleted. So is it optimal to consume 300g to 400g of carbs a day while sitting behind a desk, I don't think so. Can one consume that much if they are working out intensely at night, yes and their performance should be well fueled.

Looking at another area, if looking for fat loss/weight loss. Make sure that you are still meeting your protein requirements and then just eat less total fats and carbs. Carbs can be preferred as long as you are using the glycogen that you are storing through working out or by doing intermittent fasts (even then you have to be careful to not top off your glycogen storage tanks).

What do you think of this approach?

Aaron said...

My argument DOES NOT imply that carnivorous animals don't use glycogen. They generate it from amino acids, just as humans can. They differ from us in their tolerance for ammonia generated in gluconeogenesis.

A very profound statement. I have been arguing for years within keto circles that it's probably best to raise carb intake "at least" the point where you don't need much protein to convert to sugar. Most studies seem to show longevity benefits to a lower protein diets (esp low in sulfer amino acids)- very hard to fight the data.

At some point soon I plan to start a blog to tackle theoretical nutrition. Asking how things like carb/intake effect our hormones and how that might effect our longevity. Questions of AGEs and mineral sufficiency will also be on the table. There are many reasons why I feel tubers are a great food source.

One question: When can we grow sweet potatoes with low beta carotene content? They would be the almost perfect food then!

Anand Srivastava said...


First Congratulations for PhD. I am not able to comment on the Blogger. I am posting here from home. I generally don't do any posting from my house. No time;-).

The skin of Potatoes would not be eaten by paleolithic ancestors as it would be burnt when cooking in the ashes.

It makes a lot of sense to peel the skin out.

Malcolm Klein said...

Agree completely with your post. Just a couple of additional things to consider that I don't see mentioned-

1)Additional evolutionary support beyond the amylase gene changes is provided by Richard Wrangham's theories on how cooked tubers may explain our rapid change in dentition at 1.8 million years ago as we were forced out into the savanna by climate changes- cooking allowed us to exploit the only readily available carbs - not much fruit out there!

2)No issues with fructose- roots are 100% glucose when digested (except for some that have been bred to be sweet, like sweet potatoes or beets)

3)Excellent alkaline/acid balance as they are basically just another form of vegetable flesh and so are very high in potassium.

4) Only negative may be that our cultivated versions may be higher in glycaemic index than the wild versions we evolved on- but this can be mitigated by limiting portions, combining with other fiber rich foods or even drinking viscous fiber fortefied drinks with the meal. Another possibility is using the dried roots as flours fiber enriched low-glycemic baked goods. (Yam flour, Cocoyam flour, and Cassava flour, for example, are readily available in African grocery stores and substitute well for wheat flour)

L said...


may be i'm misreading, but it seems like you are assuming a protein to carb conversion of 100%i don't think that's right.

Eran said...

I've seen several sources indicating the protein:glucose ratio is 1:0.7.

Dr. Eades makes a point of this in a response to one of the comments here.

Don said...

I used the rounded ratio of 1:1; it got it from McDonald's book on Ketogenic Diet and didn't want to look further at the time. If we use the 1:0.7, this increases the amount of protein needed by another 30% (i.e. one gram of protein produces only 0.7 g of glucose, because 30% of protein consists of nitrogen and sulphur and ketogenic aminos that don't convert to glucose). To get 133 g glucose, you would need 190 g protein.

Don said...


"...why are so many people gaining weight with tubers who eat primal, and why are so many people fat?"

I could as easily ask, why are there people commenting on my blog abot how they haven't lost weight despite eating little or no carbohydrate?

Or, why are !Kung, Pygmies, and Hadza all lean, despite regularly consuming starchy tubers?

Answer: Calories count. If an object is gaining mass, the increase had to come from some input (conservation of matter/energy), i.e. food. You can gain weight on a low carbohydrate diet, if the energy intake exceeds energy expenditure; and you can lose fat on a moderate carbohyrate diet, if energy expenditure exceeds energy intake. Not even Taubes disputes this.

The question is, how to regulate hunger. If eating some type of carbohydrate increases your appetite, cut it out; if adding some type of carbohyrate reduces your appetite, leave it in.

Don said...


If reducing calories, at this point I would keep protein intake at the high end of the range (say 1.4 g/kg/d), because needs increase in a calorie deficit situation, studies show this reduces lean mass loss (compared to lower end of range (e.g. Layman's work), and because PRO induces satisfaction, making it easier to stay in caloric deficit.

BTW I would count only PRO in animal products, or at most, 50% of that in plants. Plant proteins have poor digestibility. Having lived through a trial of veganism, I don't believe plant proteins contribute substantially to human nutrition, due to their being bound in indigestible fiber complexes.

"...then just eat less total fats and carbs. Carbs can be preferred as long as you are using the glycogen that you are storing through working out or by doing intermittent fasts (even then you have to be careful to not top off your glycogen storage tanks)."

I agree, with the caveat that in caloric deficit, one must eat adequate fat to control hunger. I suggest that for most people, not less than 30% of calories from fat.

Don said...


Have you not seen white-skinned, whitish-yellow-fleshed sweet potatoes, developed in Japan? Delicious and much less carotene than the others.

Don said...


Wrangham's hypothesis doesn't support my suggestions, simply because it is a hypothesis, not fact...

I'm not convinced the glycemic index of individual foods is all that important. The glycemic load, yes...and tubers have a relatively low glycemic load. Therefore, I don't presently think adding fiber to tuber starch will be as beneficial as simply eating them with protein and fat, which will reduce the glycemic response without causing all the intestinal problems produced by excessive fiber.

Don said...


Welcome back. I agree. Paleopeople would not have eaten the skin of tubers for the reason you gave. Also, we don't eat the skin of bananas or citrus, why eat the skin of a tuber if we recognize that it tastes bad and gives us (me) gas?

Aaron said...

I think I've seen some white sweet potatoes once in a while at a specially store I shop-- I think they are still a rarity.

I'd love to open a business that supplies low anti-nutrient, low carotene sweet potatoes-- the market is wide open!

BTW your article also shows people why they should eat intermittently.

I suffered a separated shoulder 10 days ago and started to include way more carbs from tubers to increase healing speed-- from what I can tell- I'm healing faster than I would from my usual low carb fare (this is my personal experience)-- black and blue marks are becoming yellow faster also. I'm very pleased.

Many in the low carb community have blogged about slower healing times from injuries-- maybe its time to celebrate the carbs (as long as it isn't frutose-- and isn't in excess of your glycogen stores<----- and even then it's not that much of a problem if you can manage to keep your weight low.

Don said...


My local natural food market almost always has the white sweet potatoes. When I lived in Toledo, Ohio, I could get them at Kroger's very often.

Interesting about the healing. Some repairs depend on glycosaminoglycans.

To make them you need a surplus of glucose.

Aaron said...

Don, the only reason why I would eat the white one is for the reduced beta-carotene content-- do I really need to worry about this?

I've seen conflicting things about the yellow/orange tint people get from carotenes.

mario_encinias said...

Aaron makes an interesting indictment of beta carotene. Elighten me, as I'm sure many readers are not aware of any caveats concerning this phytonutrient.

On an aside, when I first took on the plaeo lifestyle I followed Cordain's version and saw spectacular results for the first two weeks. From that point foward I experienced persistent IBS symptoms and waxed nostalgic about those first two weeks. It took me a long time to figure out (and justify in my mind) that tubers were indeed part of our deep ancestral past, and it was until I incorporated yams and potatoes that I saw relief from the IBS.

Don said...

Aaron and Mario,

I don't know of any reason to avoid food-source carotene complex. When it accumulates in the skin, this protects against photodamage from sunlight (one reason green plants don't suffer from photodamage). On the other hand, synthetic or isolated beta-carotene (separated from its natural complex) may do harm. A 2004 study of antioxidants and GI cancers found a 30% increased risk of death in people supplementing ß-carotene plus vitamin A, 10% increase in people using ß-carotene and vitamin E.

Isolates do damage by causing imbalances.

Aaron said...

Don, check this thread:

A poster named krillin posted a study that I can't seem to load that talks about the sweet spot for beta carotene around 4000-60000 iu

what if I were to eat 4 cups of sweet potatoes a day-- that is only 1000 calories but 200,000ius of beta carotene-- isn't that too much-- would i have to worry about any of the cod liver oil i take on occasion-- i just don't trust play anti-oxidants too much.

Drs. Cynthia and David said...

Thanks for the thought-provoking post! We don't disagree that starch sources could have provided valuable calories and enhanced survival. However, we take issue with your thesis that large amounts of protein would be required to replenish needed glycogen stores in the absence of carbohydrate consumption from starches. Your arguments appear to be based on misinterpretation of references and gross overestimation of the rate of glycogen depletion resulting from high-intensity exercise for periods of a few minutes. A detailed discussion is too long for this comment, so we have posted our thoughts on our own blog.

Vladislav said...

I general I agree with your theory about starches and their role in more effective glucogen levels maintenance.
But I would comment on toxicity in case of high protein intake. Can you provide any proven reference that protein intake levels higher than these strict values you fixed in your post are harmful for healthy humans? This is a long-time controversial topic and to my knowledge all assumptions about high protein toxicity are based on studies on subjects with already impaired liver or kidney functions. It is not quite serious to expect that our ancestors have monitored the amount of meat they consume to not eventually exceed 2 lbs daily - more likely they consumed as much as available at the moment. I would speculate that they ate excessive amounts of protein (meat, fish, whatever) after a successful hunt and there were days with little or no food at all. So I guess huge amounts of protein were (and are) not such a problem.

Methuselah said...

Hi Don,

I referenced your post in my write-up of a long mountain race I took part in recently, when in a fasted state.

I know Drs Cynthia and David have written a (very long!) post following up on yours, so I will probably ask for their views too, but in any case it would be good to get your take on the conclusions I have drawn about my own experience...


Don said...


The study referenced looked at beta-carotene supplements, with supplemental amounts over 6400 mcg associated (not causing) genome instability. Too many variables in that study, and supplemental beta-carotene is not the same as full spectrum carotene complex from foods. As I stated, isolated beta-carotene can cause imbalances by mass action and increasing need for synergists present in whole foods. Take care not to confuse studies on effects of supplements (not part of our evolutionary adaptation) with effects of foods.

Don said...

Drs Cynthia and David,

I would like to post a response too lengthy for commenting here. Just to start:

First, you did not read the Fournier paper that I cited. Sorry for some reason the link did not work. Look for this:

POST-EXERCISE MUSCLE GLYCOGEN REPLETION IN THE EXTREME: EFFECT OF FOOD ABSENCE AND ACTIVE RECOVERY, Fournier et al, International Society of Sports Nutrition Symposium, June 18-19, 2005, Las Vegas NV, USA

This paper includes this quote to which I referred:

“In fact, we store just enough glycogen to sustain our energy demands for only a few hours of intense aerobic exercise (Gollnick et al., 1973; Ivy, 1991), and so little glycogen is stored in our muscles that close to a third to half of these stores can be depleted within a few minutes of a maximal sprint effort (Gollnick et al, 1973; Fairchild et al., 2003). As a result, active individuals are at increased risks of experiencing a fall in their ability to engage not only in intense aerobic exercise (Ivy, 1991), but also in short sprint effort under situations eliciting fight or flight responses (Balsom et al., 1999; Fournier et al., 2002).”

I did not misinterpret anything, I merely paraphrased the second half of the first sentence of this quote. You can contemplate the second sentence, which refers to people eating mixed diets, while I prepare a further response forthcoming within a few days.

Don said...


I will comment as soon as possible.

Don said...


Rudman et al, Maximal Rates of Excretion and Synthesis of Urea in Normal and Cirrhotic Subjects, J Clin Invest. 1973 September; 52(9): 2241–2249.

To quote Cordain, using the known maximal rates of urea synthesis [65 mg N/h - kg (body weight )], “The mean maximal protein intake for the average weight U.S. male (189.4 lbs ) is then 270 g/day (range 233-322 g/day), and for an average weight female (162.8 lbs ), 246 g/day (range 208-288 g/day).”

[From Cordain L, The Evolutionary Basis for the Therapeutic Effects of High Protein Diets. Published in The Performance Menu as part of The Protein Debate between Cordain and T. Colin Campbell. I think you can still find this online.]

Observers recorded hunter-gatherers refusing game that they had killed, but turned out to have too little fat. Cordain et al [The paradoxical nature of hunter-gatherer diets:
meat-based, yet non-atherogenic, European Journal of Clinical Nutrition (2002) 56, Suppl 1, S42–S52] noted:

“Hunter-gatherers tended to shun very small animals or fat-depleted animals because of their excessive protein content (Noli & Avery, 1988; Speth & Spielmann, 1983; Speth, 1989), and numerous historical and ethnographic accounts have documented the adverse health effects that have occurred when people were forced to rely solely upon the fat depleted lean meat of wild animals (Speth & Spielmann, 1983). Excessive lean protein consumption without adequate fat or carbohydrate causes a condition referred to as ‘rabbit starvation’ by early American explorers
that results in nausea, diarrhea and eventual death (Speth & Spielmann, 1983).”

Speth and Spielman ( (Choose the second link for the PDF) noted:

“We became interested in this situation when our research on late prehistoric and early
historic adaptations in the Southern High Plains indicated that hunters and gatherers were adopting a variety of subsistence strategies in the late winter and spring that avoided the consumption of lean bison meat (Speth 1983; Spielmann 1982).Fat-depleted animals were avoided and lean cuts of meat were abandoned by hunters at spring-season kill sites despite the facts that spring was generally a time of food scarcity in the region and bison was the only readily available subsistence resource.”

Speth and Spielman go on to note that the fat content of ungulate game reaches its lowest during winter and spring, when human caloric requirements would reach a zenith (due to the increased difficulty of hunting in snow, etc.).

So hunters who did not have an auxiliary (animal-independent) fat or carbohydrate source wasted energy killing animals they could not and refused to eat because too lean. Well established, not hypothetical.

“I would speculate that they ate excessive amounts of protein (meat, fish, whatever) after a successful hunt and there were days with little or no food at all.”

Interesting speculation…if so, how many modern people attempting to emulate primal diets are eating meat in large amounts only every few days?

Drs. Cynthia and David said...

Hi Don,

Thanks for the clarification on the reference. I see your point, but I think I still disagree with their conclusions. We should have said that they misinterpreted not you! We'll take a look at it and maybe follow up on why or why not.


Dr. Curmudgeon Gee said...

awesome post.
(taro is my favorite tuber.)

Anonymous said...

It’s always good to learn tips like your posting. I just started posting comments for blog and facing problem of lots of rejections. I think your suggestion would be helpful for me. I will let you know if its work for me too. Thanks and keep posting such an informative blogs. You may know me by back pain