Sunday, August 30, 2009

Primal Potatoes, part 1

Although in overall form the human gut appears distinctly carnivorous, humans have one unusual feature for a carnivore: we have salivary glands that produce amylase, which has the the sole function of digesting starch. Since meat contains essentially no starch, this feature could not have arisen as an adaptation to an exclusive meat diet. It clearly represents an adaptation to starch consumption. Did this arise recently, in response to the Neolithic shift to cereal-based diets, or does it have deeper roots?

Meet AMY1

AMY1 is the gene that codes for the production of salivary amylase. Perry et al compared the AMY1 copy number variation humans from low-starch diet populations, humans from high-starch diet populations, and wild-born chimpanzees. They also examined the amount of nucleotide sequence divergence among the three AMY1 gene copies found in the human genome reference sequence.

The high-starch diet populations included Hadza hunter-gatherers and two agricultural groups, European Americans and Japanese. The low-starch diet populations included Yakut pastoralists and two rainforest hunter-gatherer groups, Biaka and Mbuti. The high-starch diet people carried a median of 7 and a mean of 6.72 copies of AMY1, in a range of 2 to 15 copies. The low-starch diet people carried a median of 5 and a mean of 5.44 copies of AMY1, and exhibited a range of 2 to 13 copies. Thus, AMY1 copy number correlated positively with diet.

When Perry et al examined 15 wild-born chimpanzees, they found evidence of only two diploid AMY1 copies. Thus, whereas the humans from high- and low- starch diet groups vary but the average human, regardless of high- or low-starch diet, has about 3 times as many copies of this gene as the chimpanzee. As a result, humans produce 6 to 8 times as much salivary amylase as chimpanzees.

Perry et al found a low amount of nucleotide sequence divergence among the three AMY1 gene copies found in the human genome reference sequence, which “implies a relatively recent origin that may be within the time frame of modern human origins.” Assuming that humans and chimps diverged from a common ancestor about 6 million years ago, they estimate that the increase in AMY copy number occurred within the last 200,000 years, the time period during which modern humans (H. sapiens sapiens) emerged.

This strongly suggests that starchy food has had a significant role in human diets for substantial time periods, starting in Paleolithic times. Most likely, this adaptation arose as an advantage in digestion of tubers, corms, and bulbs. It implies that these foods played a large enough role in human evolutionary diets to give a selective survival advantage to individuals who had extra copies of AMY1 (compared to chimps) and hence an enhanced ability to digest starch. I see no other way of explaining our unusual production of an otherwise useless salivary enzyme.

As I noted in The Garden of Eating, in 1996 anthropologist Melissa Darby, M.A., of Lower Columbia Research and Archaeology (Oregon) demonstrated that Northern Hemisphere paleolithic humans had access to arrowhead (Sagittaria latifolia) a prolific wetlands plant that produces a tuber very similar to the white potato. Pollen data indicates that this tuber —called “wapatos” by the Chinook tribe – thrived in the last Ice Age throughout North America, the North American Great Basin, Siberia, and Northern Europe – overlapping the time that during which Perry et al estimate that humans started showing extra copies of the AMY1 gene.

Darby has harvested approximately 5,418 calories per hour gathering wapatos from a knee-deep pond. The tuber is most abundant from late fall through spring, when other plant foods are scarce. People can eat these tubers without grinding or mashing, and they cook thoroughly in a bed of hot ashes in 10 minutes, no oven required. They keep fresh in a cool place, and also dried.

American Prairie Indian women also gathered and cultivate the starchy camas bulb. Darby says that a woman gathering camas could net 5,279 calories per hour. This would consist primarily of starch: nearly 1000 g of it.

Thus, we know that humans had access to tubers—I’ll call them primal potatoes—even during the Ice Ages. It seems fairly certain that we have descended from a long line of tuber-eaters extending back at least a quarter of a million years. I have more to say on this in my next post.


Aaron Blaisdell said...

Interesting post. Potatoes (both white and sweet) are one of the few starchy vegetables that I still eat fairly regularly. I love sweet potato fries and white potatoes baked in the oven drizzled with lots of good fat (either beef tallow, butter, bacon fat, or coconut oil) and a sprinkling of herbs. Potatoes are a great accompaniment to any meats, but especially to the slow roasted or braised meats due to the reduction sauce. I'm getting hungry!

Anonymous said...

exactly how similar however would the tubers found then be to the ones now. they were most likely smaller, more nutritious and eatten when meat was not available, as well as not favored. they provided no fat to a hungry person which we all know wont satisfy anyone.

i cant be convinced tuber or potatos of any sort played a role in those time periods.

Mark said...

So do you think that including potatoes (sweet and white) in the appropriate amount daily would cause much harm? I know that glucose is toxic but in the appropriate amount I theorize that a spike in insulin and blood sugar is ok once or twice a day if it's coming from natural stuff like tubers and maybe fruit. What do you think? Thank you for a great blog, I reference your fiber post constantly! I refuse to believe that we have to eat veggies to alkalize our blood. Thanks again, Mark.

Don said...


Well, then, its up to those who deny that our ancestors ate starch to come up with some other explanation for the increase in copies of AMY1 and production of salivary amylase in humans compared to chimps. It sure didn't happen by pressures exerted by eating meat, since we don't need that enzyme to digest protein or fat. To get this change, something had to favor the reproduction of humans who could produce higher amounts of amylase, and the pressure had to be constant. I will give my take on it in the next post, but as a hint, you assume tubers and meat are mutually exclusive components of the diet (if they had meat, they wouldn't eat tubers). I disagree.


Short answer, I agree with you, assuming a primal meal schedule. More in upcoming posts.

Anonymous said...

there was some research published recently that showed that high glycemic foods (i.e. potatoes) interfere with the function of endothelial cells for up to several hours, in addition to making your arteries slam open. Both of these things can spell trouble for your heart.

Mark said...

Thank you, looking forward to the next post.

Don said...


Kitavans eat sweet potatoes every day, as their main food, with essentially no heart disease, despite high rates of smoking.

Anonymous said...

Kitavans eat a diet of root vegetables, coconut, fruit, vegetables and fish

if you would like to dis-clude red meat from your diet and rely on fish be my guest...that is why the diet works for is very low fat, something i doubt mark promotes

Bryce said...


Great post again. Ever since you mentioned recently that you were going to discuss the benefits of sweet potatoes as being possibly anti-diabetic, I've been checking back every day. I'm such a dork.

In any event, I'm looking forward to your next post. Hopefully it will explain if/how some tubers, especially sweet potatoes, can be eaten without racking our bodies with an insulin spike.

Thanks Don,

Bryce said...

Kudos on another good one. I was certainly curious when you mentioned a post discussing the healthy nature of tubers, and thus far you've done an excellent job making their case.

I look forward to the next installment! Also, a colleague of mine just told me she is entering a new profession: acupuncture! With any luck, I'll get her to visit your site so she can learn about appropriate diet/nutrition as well.

Don said...


What evidence do you have proving that people can't eat tubers and red meat and maintain good health?

What evidence establishes that meat and tubers are mutually exclusive in designing a healthy diet? Or that they were mutually exclusive in ancestral diets?

"...they provided no fat to a hungry person which we all know wont satisfy anyone."

Holt performed a study to determine the satiety value of various foods. 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.

Regarding the size of wild tubers, people have harvested wild Ipomoea species (same family as sweet potato) the size of a human head.

Anonymous said...


Special consideration must be given for diabetics here (like me). Sweet potatoes have an adverse effect on my blood sugars very quickly. Not all diabetics respond to the same degree to the same foods. For someone who is healthy, sweet potatoes may or may not be an issue. One advantage i have is that I can test any time I want and see the effect of different foods on my body. thanks for your blog.

Don said...


In this post (and my entire blog) I search for greater understanding of primal diet and lifestyle. I don't have attachment to any one perspective, and I won't ignore anything I know is true (such as the fact that we produce amylase). Ignoring facts is maladaptive. I look at human anatomy and physiology and various lines of evidence for the purpose. If we have a gall bladder, this indicates one thing, if we have amylase, this indicates another.

Also, I'm not posting only for diabetics, or only for obese people. My intent is to help everyone understand the human evolutionary diet.


I am not trying to convince anyone to eat sweet potatoes against his/her own best judgement, nor do I mean to say that everyone "should" eat tubers. I do have experience with seeing diabetics get their sugar under control while eating sweet potatoes, following the general guidelines in my book, but that might not work for you.

I do want to mention that even a nondiabetic will see blood sugar go into 140 or 150 range after eating a high carbohydrate but relatively low glycemic index food like sweet potatoes. This is "normal" response to eating carbohydrate. White bread will put a non-diabetic's blood sugar up to 250. How quickly it gets cleared depends largely on whether or not you have muscle cells hungry for glycogen replenishment (i.e. insulin sensitive). I don't know that we have any evidence that we must keep our sugar under 120 at all times to secure health.

Of interest, to test for diabetes we always use fasting blood sugar, but this may not be wise. It appears that diabetics can have high CREB activity that increases liver gluconeogenesis during fasting. Thus, people with high CREB activity will have high fasting blood sugar. I used to puzzle at some people who had high fasting blood sugar despite low carb diets until I learned this. The question thus arises, is high fasting blood sugar really a good way to identify diabetes, or does it also identify people who are NOT diabetic but simply have greater than average gluconeogenesis ability?

More in my next post.

Aaron said...

love this post!

are you softening your viewpoint on carbs?

I still think there may be some longevity benefits from the lessening of mitochondrial byproducts by relying on some energy from carbohydrate.

have you looked at this type of info:

potatoes seem to be a clean fuel-- low anti-nutrients, no fructose, low protein, has minerals and vitamins

if we can prove less mitochondrial stress from carb intake vs fat-- insulin may not matter as much if u stay thin. we already know that protein revs up mito stress!

Anonymous said...

Thanks, Don

Jonathan said...


Check out these two posts from Peter regarding mitochondrial stress from glucose vs. fatty acids:

In short, keytones and fatty acids appear to be better for mitochondrial health than glucose.

Don said...


interesting abstract on protein/methionine restriction, especially the mention that such activates the SIRT1 protein. It mentions that neither carbohydrate nor fat restriction reduces ROS generation, so doesn't support the idea that switching fat for carbs will reduce mitROS exposure. But it does look like subbing carbs for protein would do so...which does dovetail with my next post on the adaptive advantages our ancestors realized by accessing tubers.

Marnee said...

Or the production of amylaze is an adaption that is on it's way out. How does one determine that? I don't think you can, necessarily but I can speculate. The human digestive system does not react well to starches, and sugars are especially damaging despite saliva, not to mention the higher incidence of dental carries among carb eaters. One trait working against lots of others. Mmm hmm. Just because you can eat it and live doesn't mean ya should.

Aren't wild tubers kinda toxic?

"Wild yams make a significant contribution to diets of tribal people in Nepal. However, these wild tubers are unpalatable, taste bitter, produce inflammation and show occasional toxicity."

"Root crops, in common with most plants, contain small amounts of potential toxins and antinutritional factors such as trypsin inhibitors. Apart from cassava, which contains cyanogenic glucosides, cultivated varieties of most edible tubers and roots do not contain any serious toxins. Wild species may contain lethal levels of toxic principles and must be correctly processed before consumption. These wild species are useful reserves in times of famine or food scarcity. Local people are aware of the potential risks in their use and have developed suilable techniques for detoxifying the roots before consumption."

Don said...


"...not to mention the higher incidence of dental carries among carb eaters."

I will address this in one of the few posts I have on this topic....

FAO says:
"cultivated varieties of most edible tubers and roots do not contain any serious toxins."

Most likely these came from wild stocks that had low toxicity.. natural selection at work (that's what humans would naturally select).

In other words, the still wild ones you point to are the ones that didn't get selected by humans... so no surprise they are (still) toxic.

The genetic data definitely don't support the idea that the amylase adaptation is on the way out. Chimps have less, we have if anything it would be "on the way out" in chimps, not humans.

Drs. Cynthia and David said...

It makes sense that there would be a survival advantage in being able use tubers to supplement calories in addition to animal foods, but it doesn't necessarily imply optimal health would come from eating them, or what amount would be optimal if they are eaten. That said, I eat potatoes now and then in small amounts, particularly after exercising a lot (having earned my carbs and I know my muscles are eager to take up glucose), but I'm still trying to lose weight. It might matter less for people just trying to maintain weight.
I suspect whether tubers are a good addition to diet depends on quantity and each individual's insulin sensitivity, etc.

Thanks for the thoughtful post.


Anonymous said...

I would note two things. First, the 200,000 year limit is a maximum; this may well be a neolithic adaptation rather than a paleolithic adaptation.

Second, it's known that potato starch in particular is considerably less digestible than corn and wheat starch; see Langworthy and Deuel, Jr, "Digestibility of Raw Corn, Potato, and Wheat Starches", Journal of Biological Chemistry, 1920 (the article doesn't have the journal issue on it but says "received for publication February 26, 1920).

If we did eat tubers - which I question, as I would lean towards this being an agricultural adaptation - that doesn't mean potatoes are okay.

Don said...


If neolithic, how come it appears in hunter-gatherers having no history of agriculture?

"Digestibility of Raw Corn, Potato, and Wheat Starches"

By 200K YBP, people used fire. I don't see the relevance of a study of the digestibility of raw potato starch.

Anonymous said...

"If neolithic, how come it appears in hunter-gatherers having no history of agriculture?"

Because of normal human gene flow. In my opinion, humans with more than two copies of the amylase gene likely appeared first in agricultural areas where the selection pressure for those mutations would have been huge. Those mutations would be nearly selectively neutral in other human populations, and would randomly diffuse throughout the world over thousands of years. In that one hunter gathering group ready access to starchy foods, the chromosomal alleles with more copies of the amylase gene would be concentrated.

I say "more than two copies" because it looks to me, based on the data in the paper, that the third version of the amylase gene didn't appear until only about 25,000 years ago. That would then be a rough upper limit for more than two copies of the amylase gene, as we would otherwise expect more different versions of the gene. Given that all humans appear to have at least two copies of the amylase gene, we could possibly have had exactly two copies of the gene as far back as 200,000 years - but it's hard to believe that substantial consumption of starchy tubers would have selected for a second copy of the gene but no more. I think a more likely explanation would be use of foods with low levels of starch, such as perhaps root vegetables that are not tubers, like carrots.

With respect to the potatoes, I question whether cooking in the embers of a cooking fire would have uniformly cooked all the starch. Even a small amount of uncooked potato can cause lots of painful gas, as I can personally attest to. Potatoes, being a new world food, were not available to humans until the neolithic; it would be interesting to see if there are similar digestibility concerns for yams, which would have been available to humans in the paleolithic.

Christoph said...

"i cant be convinced tuber or potatos of any sort played a role in those time periods."

Then you're mistaken.

I lived off the land on Vancouver Island (rain forest ecosystems, mostly) as a teenager. Tubers were a big part of my diet... as were vegetation including greenery and cambium, fish, insects, and occasionally small game I trapped or hunted.

I assume (and know for a fact) the coastal native peoples of BC were better hunters and especially fishers than I was. They undoubtedly secured more calories from the environment.

But if you think rhizomes and tubers weren't a significant part of their diet, you're simply wrong.

Read that again.

And as to the idea they weren't "favored", I'd say that depended a lot on the individual then as it does now.

Of course it doesn't have fat. They would, when possible, eat it with other foods that had fat and protein.

Just like we do now!

Don't eat them if you don't want to. I'm avoiding them for the most part.

But to say that despite all available evidence, "I can't be convinced..." seems to be more a flaw in your thinking than a defect in the evidence.

You're entitled to your own opinion. But not your own facts. The facts support that humans in many different populations have eaten starchy food for tens of thousands of years.

Just as you're entitled to your own opinion, you're also entitled to change your opinion. Never forget the fact.

I would never say potatoes are as good a food as salmon (or yummy yellow-orange salmon berries). I don't think any coastal native people would.

But they certainly ate both.

And while I don't have a copy of it anymore, when I lived off the land I had a handbook of foods used by the coastal native people's of British Columbia, and how they found them.

I assure you I wasn't the one to come up with the ideas of using tubers. I read about it.

Finally, starch isn't damaging to the body in the same way excess fructose is (and you wouldn't deny early humans ate sweet fruit when they could find it.)

This seems to me to indicate, if anything, we're better adapted to larger amounts of our calories from starch than fructose.

Chew on that.

Christoph said...

I'm in the same boat (unfortunately probably a "bigger" boat).


I agree with your thinking on starch explicitly. I thought you summarized everything nicely and appreciated your comment.

Christoph said...

"...starch isn't damaging to the body in the same way excess fructose is..."

And you're making me rethink some of my thoughts, Don.

I should think scientifically and follow the evidence where it leads (and not just admonish others to do so!).

Anonymous said...

It is very nice and helpful information, success for you.
If you are mind, please give me a comment or advice on my blog, an English language blog. A billion thanks for you. back pain

Anonymous said...

Hi, good post. I have been thinking about this issue, so thanks for sharing. I will definitely be coming back to your blog. back pain

Lake McCarthy said...

I have been looking for this post for awhile. I could never wrap my head around the vilification of all types of potatoes in the diet. It just never made sense from any angle, including Paleo. After eliminating grains, I saw my energy completely nosedive even after fat adaptation. The answer for me was not in adding more dietary fat but instead carbohydrate. I started looking more and more to tubers as these always made me feel the best. Every 48-72 hours dependent upon activity level, I need to refeed with something along the lines of a small to medium potato and also add in a low glycemic load starch such as squash or pumpkin a few times a week to keep energy levels constant. Without the starch, I crash and suffer relentless brain fog. I also drop weight too quickly and sacrifice muscle mass. Everything you explained lines up with my own experience. Thanks so much for this post.