Vegetable oil fatty acids are not essential. 

They are conditionally essential at best, only if docosahexaenoic acid (DHA) is lacking.  We can’t synthesize omega 3 fatty acids, and indeed they do prevent/cure certain manifestations of “essential fatty acid (EFA) deficiency” (Weise et al., 1958), but DHA can do all that and more.  Not that I recommend this, but a diet completely devoid of 18-carbon vege oil fatty acids will not produce EFA deficiency in the presence of DHA. (“vege,” rhymes with “wedge”)

Essential fatty acid metabolism


The “parent essential oils” are linoleic acid (LA) and alpha-linolenic acid (ALA).  The others, which I think are more important and the truly “essential” ones are eicosapentaenoic acid (EPA), arachidonic acid (AA), but mostly just DHA.

The first manifestation of EFA deficiency is dermatitis (Prottey et al., 1975).  Some people say LA is necessary to prevent this, but it would be better phrased as “LA prevents dermatitis;” not “LA is necessary to prevent dermatitis.”  All of the evidence suggesting LA is essential is in the context of DHA deficiency.

Technically, we can convert a bit of ALA to DHA, estrogen helps, testosterone doesn’t (women have better conversion rates)… and I’d speculate that the reverse is probably easier (DHA –> ALA).

Exhibit A.  Feeding mice a fat-free, high carbohydrate diet induces marked microvesicular and macrovesicular hepatic steatosis rather quickly (Alwayn et al., 2004).

fatty liver steatosis

A, chow (liver fat = 3.4%)
B, fat-free diet (24.1%)
C, supplemented with purified fish oils (7.2%)
D, supplemented with soybean oil (8.1%)

This study showed DHA was at least as good as vege oils in this context…

Exhibit B.  EFA deficiency, not prevented by a diet with zero polyunsaturated fatty acids (PUFAs, eg, LA, ALA, DHA, etc.)… fat source was hydrogenated coconut oil (HCO), and EFA deficiency was fully prevented with 4% DHA + 0.2% AA.

Essential fatty acid deficiency

HCD-only: fat-free, high carbohydrate diet, just like in exhibit A.
HCD-0: high carb diet supplemented with 5% HCO. It didn’t prevent EFA deficiency.
HCD-0.21: high carb diet supplemented with 0.2% DHA and 0.01% AA. It almost prevented EFA deficiency.
HCD-2.1: supplemented with 2% DHA and 0.1% AA.  It worked.
HCD-4.2: 4% DHA and 0.2% AA.  It worked.

And, similar to exhibit A, steatosis was completely prevented when the diet contained at least 2% DHA (and 0.01% AA):

Steatosis prevented with DHA

^^^that’s “acute.” Here’s “chronic:”


Steatosis DHA 19 weeks


Now we’re getting somewhere.  In the beginning, we thought LA and ALA were essential.  Then we thought EPA, DHA, and AA.  Now it’s just DHA and AA? …the top tier omega 3 and omega 6 fatty acids, respectively.

Exhibit C. AA isn’t even “essential” in a diet with no vege PUFAs, supplemented with EPA and DHA (Ling et al., 2012):


Essential fatty acid deficiency purified fish oil


So, the diet needs either DHA and EPA or DHA and AA.

Exhibit D.  DHA alone (Le et al., 2013)


Mice were fed, for 3-5 weeks, diets with 2.1% DHA plus AA, in ratios of 1:1, 5:1, 10:1, 20:1, 200:1, and 100:0.  Yes, “100:0” means DHA alone.  Do not pass GO!  Do not collect $200.  No vege PUFAs, EPA, or AA.

These were synthetic diets, with purified esters of DHA and AA.  Clear biochemical evidence of EFA deficiency, completely abrogated by DHA.

No steatosis with DHA alone 1

a closer look:

No steatosis with DHA alone 2

I propose a new meaning for EFA: DHA.

I know LA and ALA do stuff in the body, and it’s virtually impossible to construct a diet with ZERO LA and ALA.  But: 1) providing all of the fat as coconut oil and purified DHA comes pretty close; and 2) as mentioned above, your body can probably make some ALA/LA from DHA/AA.  And there’s no need to avoid LA and ALA, I’m just saying it’s incorrect, or at least incomplete, to say they are the essential ones, when the long-term consequences of their deficiency pales in comparison to that of DHA.  Don’t get me wrong, dermatitis sucks, but compared to impaired cognitive function and debilitating mood disorders, itchy skin is a walk in the park.  (although evidence suggests it’s not an either/or situation; DHA prevents all of the above whereas LA only prevents dermatitis)

Remember Dr. Eades’ epic 2007 post about the Great Starvation Experiment?  I’m pretty sure Sam Legg went nuts and chopped off his fingers in part because of DHA deficiency.

Part II: seafood > fish oil supps >> no DHA

DHA from seafood is likely more effective than that from supps, but supps aren’t worthless.

However, many fish oil pills contain unacceptable levels of potentially adverse peroxidation products.  Out of 63 different products tested, Rupp and colleagues showed that many of ‘em did.  Store in the fridge and discard if stinky.

unacceptable levels of potentially adverse peroxidation products

unacceptable levels of potentially adverse peroxidation products


Lastly, to get the most bang for your buck, it’s not as simple as “wild vs. farmed,” see the Fish Blog to see how DHA & contaminants vary by geography.

As Dr. Kruse has been saying, DHA must be present at the “sn-2” position (that’s basically the middle of a triacylglycerol or phospholipid).  This is how DHA is present in seafood but not always in fish oil supps (eg, Litchfield 1968 and Ando et al., 1996).  It’s not all-or-nothing, but it’s safe to say you’re gonna get more sn-2-DHA in seafood.

Why “sn-2?”

1) DHA is more stable in this position (eg, Wijesundera et al., 2008).

2) It’s more bioavailable here (eg, Christensen et al., 1995).

3) sn-2-DHA is more effectively incorporated into brain lipids (eg, Thies et al., 1994 and Lagarde et al., 2001).  Again, it’s not all-or-nothing, but more is likely gonna get in your brain if you eat seafood.

Do not, under any circumstances, feel the need to supplement your diet with LA, ALA, or vege oils.  Just eat seafood.  DHA supps are negotiable (they’re better than nothing, imo).

calories proper

If you like what I do and want to support it, check out my Patreon campaign!

UPDATED Affiliate links: still looking for a pair of hot blue blockers? Carbonshade and TrueDark are offering 15% off with the coupon code LAGAKOS and Spectra479 is offering 15% off HERE.

If you have no idea what I’m talking about, read this then this.

20% off some delish stocks and broths from Kettle and Fire HERE.

If you want the benefits of  ‘shrooms but don’t like eating them, Real Mushrooms makes great extracts. 10% off with coupon code LAGAKOS. I recommend Lion’s Mane for the brain and Reishi for everything else.

Join with this link

Be Sociable, Share!
  • Melissa

    What do you think about how FADS variations might play into this? Anecdotally I had acne when I restricted LA and ALA. I’m certainly not restricting them now and my skin is clear again. I suspect I get most of this kind of thing from animal fat, particularly pork, though.

    • Interesting.

      There was a study about a decade ago that identified a population with a particular FADS variant… total n3 intake wasn’t correlated with cardiovascular outcomes but marine n3 was. So it was like, in this population, seafood or DHA supps were mandatory.

  • Tuck

    You rebel, you. 😉
    So do you think all cases of steatosis are the result of DHA deficiency?

    • there are many causes of steatosis, eg, alcohol, insulin resistance (&/or diet), some drugs, etc…

  • Tuck

    The Eades link does not work btw, this one (seemingly identical) does:

  • Michael

    I agree about the fish oil supps and never take them. Apart from shellfish I like salmon. But the fresh farmed salmon that seems to be ubiquitous here is very fatty compared to the wild, unpleasantly so, as are most farmed fish species. So I actually prefer canned salmon, which is wild, has skin and bones, etc. But I have often wondered about the extent of lipid oxidation in canned fish. Any idea whether this would be of any concern?

    • I don’t know of any studies that tested it directly, but my guess: canned fish is better than supps.

      As to the wild vs. farmed issue, check out the Fish Blog 🙂

      • This Old Housewife

        I once saw a documentary on how farmed fish get fed–the get pelleted grains, which will mess up their Omega-3 in the meat, so what’s the point of eating it to gain Omega-3 in the diet?

    • This Old Housewife

      If only Hubby could eat fish…and like it. He can eat shellfish and squid/octopus, but it raises his BG too high. That’s why we do supplements.

  • Tuck

    I’m also curious about your thoughts on this:

    “LA competes with alpha-linolenic acid (18:3n-3) for endogenous conversion to the LC derivatives EPA and DHA, and LA also inhibits incorporation of DHA and EPA into tissues. Thus, high-LA levels in the diet generally result in low n-3 LCPUFA status.”
    Ray Peat may send you a thank-you note. 😉

    • I believe it’s true, but not the whole picture: it’s hard to tease out the importance of total n3 intake from n3/n6 ratio… I think both are important, but I wouldn’t rely on either to get my DHA… the conversion is pretty crappy

  • Kindke

    Bill one of your best posts,

    I liked the sn-2 part as i didnt know that was important for DHA although I have seen in other situations that the sn position in the triglycerides can greatly impact the metabolism of fatty acids and their fates.

    I haven’t researched it at all but I do wonder if cooking or certain types of cooking destroy DHA in seafood?

    • Thanks, Kindke 🙂

      Cooking won’t destroy DHA, it’s rather stable (in sn-2)… but I wouldn’t be surprised if extended high heat exposure damaged it somehow, like with other unsaturates.

      • Jack Kruse

        Cooking does have an effect and so does freezing. Freezing can diminish DHA availability by 40%. The best way to do seafood is raw. This is why Okinawans have longevity.

        • This Old Housewife

          What about the lead?

        • Tom

          Jack, no concern about parasites? I love sushi, but (at least in the state I live in) it must be flash deep-frozen before can be sold.

          I am wondering if it might be enough to get frozen salmon fillets from Trader Joe’s or Costco – and eat those upon defrosting. But I don’t know if those are actually deep-frozen or not. I suspect not.

          • frozen seafood is better than none!

          • Jack Kruse

            none Tom. electrons turn on your T regulator cells which take care of both sides of the immune system

          • Paleo Osteo

            you dont gotta worry about an infectious agent, you gotta worry about your cells ability to deal with it

        • Jack is always making me do my homework 😛

          Indeed, freezing reduces DHA:

          Effect of frozen storage on fatty acid composition and changes in lipid content of Scomberomorus commersoni and Carcharhinus dussumieri

          Fatty Acids Changes during Frozen Storage in Several Fish Species from South Caspian Sea

          • Jack Kruse

            Bill this is no homework. the reason I respect you and Peter is because your pure to truth. We are all in this together and when you realize this we must get it right for people so they can tap the doctor inside them. Eating pills, Quest bars, bacon, and paleologix supplements is a dead end road. We must strive for fundamental truth to how we really work. I understand that the scale is difficult for some to stomach. But to get where we all want to be we need to sink to electrons and protons. And I am grateful for you Bill. You have a very open mind and Although we have never met I respect you in a deep way.

    • This study showed in raw salmon, EPA+DHA = 1200 mg. After pan-frying, down to 700 mg.
      It’s probably just in the pan 😛

  • Jack Kruse

    BOOM. I been saying it and paleo has ignored it. Grass fed cows also do not have most of the DHA in the sn-2 position in case you were wondering. OK if your not a human. But if you want optimal…….crack some oysters, raw ones!!!!

    • This Old Housewife

      more crabs, not carbs!

      Can I have this in billboard size? 🙂 Just a t-shirt will do.

      It was ingrained in me (possibly my own error here) that since the animals ate the vegetation FOR us, we didn’t necessarily need it so much—WRONGO! My husband, for example, now needs leafy greens 3x/day (in some form or other) just to keep his blood sugar under 100 (yes, I think I finally fixed him). Since he’s male, I doubt that the Omega-3 conversion process has anything to do with what’s going on here, but he does get 1gr. of calamari oil 2x/day (allergic to fin fish) just to make sure he gets something in the way of EPA.

      We’re almost working a reverse Paleo here (from what I understood): instead of loading up on proteins (grass-fed meat) and eschewing the veggies, we’re minimizing the meat (still grass-fed), and maximizing the fat (by choosing meats with more fat than protein in it) and leafy greens.

      Just last night, we had a small green salad and 2 Swaggerty Farms sausage patties (the sausage recommended by Jimmy Moore). His blood sugar went DOWN a point instead of rising, and rose to baseline in the second hour. This morning’s FBG was 77. Today’s lunch-for-breakfast: leftover salad with 2 cooked patties ground up and mixed into the salad (last night’s leftovers redone). Dinner tonight will be a wrap with (of course) green salad, shredded cheese, and a little shredded smoked pork.

      Maybe Paleo has changed now, but when I was still on the wagon, they were shreiking about bacon, bacon, bacon, and all things meat as long as it was grass-fed. Some carnivore site (Danny Evers?) was pushing an all-meat diet and how to meet RDAs without vegetation. Hubby loved the all-meat thing until his BG numbers started revealing the truth. I had read about gluconeogenesis from MDA in the past, but didn’t think it applied here…until recently. Now I know FOR SURE it applies here! Grass-fed meats aren’t enough if you’re over 50 with BG issues.

      • AlexB

        At first I thought paleo was low veg, but as I read more I discovered that wasn’t true at all. Veg has always been a very important part of paleo. Loren Cordain’s original version was “eat lean meat and veg”. Robb Wolf was “grass-fed/wild-caught meat and plenty of veg”. Mark Sisson says veg should make up a lot of the diet as well.

        It will have changed since you last looked, because it is constantly evolving, but there hasn’t been a shift from meat to veg because that was already there in the philosophy. The message is just getting out there a bit more now. It helps that there are more people with science and medical backgrounds on board and helping to improve the entire lifestyle going forward.

      • “It was ingrained in me (possibly my own error here) that since the animals ate the vegetation FOR us, we didn’t necessarily need it so much”

        lol maybe people who get more of their protein from carnivorous species need to eat more plants to compensate 🙂

    • Interesting about grass-fed beef; I didn’t know that… then again, I don’t rely on beef or plants for my DHA…

      • Tom

        Bill, how do you get your DHA?

      • rs711

        We putatively rely on land mammals exclusively – if we were eating properly fed ones, that is. [ but we don’t live in such a world, so eat your damn seafood people! Take it from someone who goes into anaphylactic shock if he throws back even a oyster 🙁 ]

        >> Effect of feeding systems on omega-3 fatty acids, conjugated linoleic acid and trans fatty acids in Australian beef cuts: potential impact on human health< an interesting aside…”There was an increase in muscle 18:1- trans FA in the STGF (short-term grain feeding) group and further increased in the LTFL (long-term feed lot) group, suggesting that length of grain feeding is probably a factor for increased trans fatty acids in meat due to an increased intake of dietary linoleic acid, the pre-cursor of rumen generated trans 18:1 FA”

        Nota Bene: Financial support for this work was generously provided by the Meat and Livestock Australia

        • all good stuff ~~~

          consistently see increased 22:6n3 in pastured beef, and while total SFAs unchanged, interesting increase in our darling 18:0 (in some but not all studies)…

          Odd that they didn’t mention which trans-18:1 was formed. Perhaps we are to assume it’s ze viper

          • rs711

            cross-fingers it’s vaccenic!

    • Michael

      Yes, I’m so sick of crab restriction.

    • Ben

      I tend to think of marginal biological effects like interet rates. Three percent compounded quarterly is not big after 3 months but it is huge after 30 years. Likewise, a three month experiment may not show a huge effect, but what will that variable do over 30 years?

    • I found some studies showing markedly higher DHA in grass-fed beef (which may compensate for less sn-2???), eg,

      but couldn’t find anything about the sn-2…

      and also more stearate in grass-fed (which I consider a plus): &

    • “It’s time for you to know that nature’s cure is to eat more crabs, not carbs!”

      that’s epic!

  • Guest

    I was just reading Chris Masterjohn’s “Precious Yet Perilous” article (, looking for references, and then saw you posted this on facebook. Awesome post, once again, Bill! That being said, it is clear, as indicated by the Burr’s original research (the researchers proving “EFA deficiency”), that EFAs, namely DHA/EPA/AA/etc. are needed as only 0.1% of total calories when liver is included in the diet. Pyroxidine facilitates various fatty acids’ conversion to EFAs, too. Also, I’m not sure why raw seafood is being stated as the key to Okinawan longevity – that’s not really supported in the (quality) literature: this, and other studies from the 40’s/50’s on Okinawan longevity, show that Okinawan’s consumed ~4.8g/day of PUFA. They had almost a 1:1 ratio of SFA:PUFA.

    Also, take a look at the healthy Maasai individuals:

    “Long-chain n-3 PUFA (EPA, DPA and DHA) made up only 0.15% of the ingested fatty acids, but 5.9% of red blood cell fatty acids.The study indicates the Maasai diet is rich in SFA and low in PUFA. Nevertheless, red blood cells are composed of comparable proportions of long-chain n-3 PUFA to populations consuming higher amounts of this fatty acid group.”

    The above pretty largely confirms what the Burrs found decades ago, and what Dr. Masterjohn reported. There is also a study in rodents which compared sucrose vs. starch based “EFA deficient” diets, and I believe some coconut oil added to the diet reveresed the symptoms of “EFA deficiency” (eczema, skin and scalp problems, etc). I think the PUFA content of the diet was like 0.015%. So, apparently for great health and cognitive performance, you only need as much DHA/EPA as you’d get from a diet containing some odd amount of milk, liver, and maybe occasional eggs.

    • thanks!
      You don’t need *much* DHA; my point was only that ALA & LA technically don’t really qualify as “essential,” at least in the classical sense.

      • rs711

        Yes. From Masterjohns stuff “there is some evidence that the body might be able to convert lauric acid to ALA” (follow ref.) ==> “In conclusion, the present results showing the ?9-desaturation of lauric acid into 12:1n-3 in cultured cells and liver
        homogenates have led us to ask whether a possible biosynthesis of “essential” FA could exist under some circumstances in animals.” [Lipids, Vol. 37, no. 6 (2002) P. LEGRAND ET AL.]

        How clever!

        • cool!

          I also speculated that DHA might be able to be “back-converted” into 18-carbon “EFAs.”

  • This Old Housewife

    Technically, we can convert a bit of ALA to DHA, estrogen helps, testosterone doesn’t (women have better conversion rates)… and I’d speculate that the reverse is probably easier (DHA –> ALA).

    What about post-menopausal women? In light of this, maybe we ought to change the USDA dietary recommendations about fish.

    • it depends on estrogen levels. HRT probably helps, too.

  • Algal-oil capsules & cooked salmon: nutritionally equivalent sources of DHA.

    Food and nutrition professionals question whether supplement-sourced nutrients appear to be equivalent to those derived from natural food sources. We compared the nutritional availability of docosahexaenoic acid (DHA) from algal-oil capsules to that from assayed cooked salmon in 32 healthy men and women, ages 20 to 65 years, in a randomized, open-label, parallel-group study. In this 2-week study comparing 600 mg DHA/day from algal-oil capsules to that from assayed portions of cooked salmon, mean change from baseline in plasma phospholipids and erythrocyte DHA levels was analyzed and DHA levels were compared by Student’s t tests. In post-hoc analyses to determine bioequivalence, least-squares mean ratios of percent change from baseline in plasma phospholipid and erythrocyte DHA levels were compared. DHA levels increased by approximately 80% in plasma phospholipids and by approximately 25% in erythrocytes in both groups. Changes in DHA levels in plasma phospholipids and erythrocytes were similar between groups. As measured by delivery of DHA to both plasma and erythrocytes, fish and algal-oil capsules were equivalent. Both regimens were generally well-tolerated. These results indicate that algal-oil DHA capsules and cooked salmon appear to be bioequivalent in providing DHA to plasma and red blood cells and, accordingly, that algal-oil DHA capsules represent a safe and convenient source of non-fish-derived DHA.

  • Dustin Sikstrom

    What does this mean in terms of human servings of recommended amounts of DHA? Our brains have a bit higher ratio of size:carcass compared to mice/rats, so I’d assume our DHA need is a bit higher than their’s. But I see a lot of the need being discussed with no reference to actual numbers for humans. I’m trying to read through Cunnane’s Human Brain Evolution, but that thing is dense and I haven’t stumbled across any numbers yet either, if he even does give comment to a number.

    Speaking of Cunnane, he seems to give a bit more emphasis to AA than you do in this post, but I guess your point is that DHA is the only one that’s essential because AA can be synthesized from DHA?

    • ~0.5% of your total calories for DHA + EPA should be enough to avoid deficiency if it’s from whole foods… 1-2 grams/d.

      I was only speculating that other fatty acids (ALA, LA, AA) could be synthesized from DHA! The data are probably out there, but I haven’t seen ’em. But in the context of this blog post, dietary AA seems optional if adequate DHA.

      • Dustin Sikstrom

        Awesome, follow-up question: if 1-2g/day (~0.5% total energy intake) is *necessary* to avoid deficiency, do you have a comment on potentially finding “optimal”? (since need != optimization, right? ;))

        My thoughts are that 2-4% of total calories is optimal?

      • Jack Kruse

        Another key is to understand the real role of EPA in humans versus other mammals. EPA is the brake on omega 6’s incorporation into CM and tissues. This is why natural seafood has the specific ratio’s it has. We lose that ratio in supplements. Just as a car needs alignment to travel properly so does our journey of success in science. If you’re not aligned with the natural principles in science you can never uncover her recipes for innovation. When we align our thoughts at the correct scale of action, proper deductions and actions will follow helping improve our knowledge. Aligning our thoughts with the natural truth compels us to be honest about our priorities; about what we have allowed to get in the way of those priorities and realign and adjust them to take the right action. Alignment is vital to personal success and scientific progress in my opinion. Consider this: think the lack of focus on DHA iodine and selenium in their biologic package is why many “keto-critics” recommendations miss the boat because they do not understand the scale of how the individual parts work in unison with the environment and cell membrane biochemistry. This is the reason why there are many people who complain of adverse “symptoms” when they go into ketosis because they are no using seafood in their template.

        • Jack Kruse

          Danny mentioned elsewhere this: I have often wondered why arguments against ketosis always mentioned thyroid health but only focused on insulin being “needed” for T4 to T3 conversion, with no discussion of other important factors related to thyroid health, e.g. iodine status.
          As I had previously thought about it, the “low-carb flu” could be down to people being metabolically inflexible. So whilst they were eating a LCHF diet, they were still using too much glucose, essentially never getting their RQ low enough to be oxidising the “correct” amount of fat. My answer was simple to Danny: T3 and T4 both need multiple iodine atoms to work. This is another reason the ancestral practitioners just dont get it. Their diet is wholly deficient in this atom when you source DHA in cows. I wrote about that in the top ten paleo supplement list for a deep reason. Moreover, people don’t understand how iodine affect hydrogen bonding networks in CSF and in intracellular water by making H+ come closer together using the Grotthuss mechanism. This increases the energy transfer capabilities in CSF and intracellular water than surrounds organelles like Mitochondria.

          • Jack Kruse

            Danny then concluded this: So, just thinking about this would I be accurate to conclude that when a ketogenic diet is combined with sufficient DHA and iodine, we don’t have the same “need” for insulin to play a role in t4 to t3 conversion, and therefore we don’t need carbohydrates for thyroid function as so many have come to recommend recently?

          • Jack Kruse

            I have mentioned many times on my blog and forums and on our FB groups that the mainstream ancestral people don’t get the ” atomic scale argument” when it comes to things like iodine and T3 and T4. One of the great things about seafood is that it contains selenium which helps convert thyroid hormones to active T3. While you can eat more carbs to increase free T3 levels (a common paleo recommendation), this increases the need for iodine even more. This creates a huge issue chronically especially in the CSF. (…/S0306-9877…/abstract). Most of the T4 is stored in our gut, so by eating seafood, you allow form to meet function as the seafood can permeate the gut where majority of T4 is stored, allowing for the conversion of more T4 to T3. While DHA has pretty amazing quantum effects as Bill has talked about here, it needs to be paired with iodine (as that’s the HUMAN body’s preferred antioxidant system) to protect DHA from oxidation. This occurs in synapses due to this Grotthuss mechanism especially for heat transfers. So, we’ve come full circle. You can either eat seafood and get a great evolutionary package of things like DHA, iodine and selenium, which will support thyroid health, or you can follow the traditional paleo path lacking in seafood DHA and iodine and continually up your carb intake to offset this loss. The leadership of paleo needs to tell people this will trigger a need for more iodine and since you’re not getting it, your body will divert to backup antioxidant systems like vitamin C, vitamin E and glutathione. NO one is saying this because it hurts their business models. Now you see why so many young paleo folks follow “their thought leaders” eat lots of carbs and take supplements like vitamin C year-round. They’re forcing their bodies to rely on a suboptimal pathway which requires this constant band-aid approach.

          • Jack Kruse

            and then this: ‘famous paleo books” tell you to take a ton of fish oil supplements and their leaders are selling the stuff. This is a sure sign grass fed meat does not cover the modern humans needs. The more technology you use and blue light you allow the more DHA is turned over in CM and tissues. Where they fail is they don’t understand that supplemental fish oil does not come in the SN2 position therefore it is problematic. Pills don’t equal oysters ever no matter how famous your diet/cook books might be.

          • Jack Kruse

            DHA turns light into an electrical signal. When DHA is missing the electrical signal is missing in CM and tissues. What are the implication of this? Mitochondria have two CM and a matrix filled with H+. This is where mass charges are saved. Without DHA you can no longer use them as a super capacitor of electric charges. Why is this big? Without a charge is alters the masses of the subatomic particles within mitochondria. This is physics 101. This also points out why modern obesity researchers fundamentally clueless about fat. None of them are aware that according to Einstein’s mass equivalence equation (E=mc2) that the mass of subatomic particles is an energenic variable that is subject to the electric charge of the surrounding environment; that is why the universal constant for gravitation has been shown to consistently inconstant by physics experiments when they have decided to measured it in space. Well that science scales to the surrounding environment in our mitochondria too. These laws are universal. No one seems to realize it. This implies that as our mitochondria lose their electric charge because DHA and iodine are missing in action, masses of all subatomic particles in mitochondria increases as well. At a macroscopic level where biology scale resides, we observe increases in our weight gain because we are losing energy from the interior of the mitochondrial matrix as ECT slows. As ECT slows electric charge drops. Hyperlipid has been pounding this point in his proton series for months. No one gets the implications. This is why we humans get fat fundamentally.

          • Jack Kruse

            The less ideal your mitochondria hold their electric charges, the higher the fat version of the keto diet has to be. This is why Jimmy Moore can’t handle carbs but most of the rest of us can. I would remind you all how does Jimmy make his $$$? TECHNOLOGY. He can never access autophagy to recycle his shot mitochondria. Guys like Ben Greenfield has bad mitochondria because of excessive endurance exercise but he can lower his fat content and use carbs because he can recycle mitochondria with a lack of electric charge. As your mitonchondria improve, the lower the fat content needs to be; as the protein levels can increase…… you improve further the protein levels in your diet should decrease because your ubiquination pathways improve so you don’t not need to replace a lot of protein in your tissues. Bill has pointed out that people who are big time exerciser’s don’t need a lot of protein. This is 180 degree opposite the paleo idea. They love bacon and crossfitting. As you improve further, you begin to be able to exercise more, building your muscle mass as well and then you can add “seasonal carbs” back to your template with out risk. This is the part the low carb people do not understand. The more you exercise the less protein you actually need. This is another thing that the paleo kids do not understand because their “leaders” have no idea how ubiquination rates are tied to the amount of exercise you do. Ask Bill Lagakos if this is bad advice. I do not want anyone buying what I am selling. I want you to all know why I say what I do. My understanding is based upon a different “scale of science” than paleo biochemistry.

          • Jack Kruse

            ^^^^this explains the mTOR controversy in aging too.

          • Guest

            This is why the followers of their dogma need to be educated. They have no clue about the scale of action of how mitochondria really work. This is why they have a lot wrong. 10702032_10203939238260096_3846484096644599847_n.jpg

          • From Jane Plain on Twitter: “[Physical] activity happens when the brain and body make energy normally.”

            Peter’s phrase “duff mitochondria” comes to mind…

          • This.
            “You can either eat seafood and get a great evolutionary package of things like DHA, iodine and selenium, which will support thyroid health, or you can follow the traditional paleo path lacking in seafood DHA and iodine and continually up your carb intake to offset this loss.”

          • more support for your seafood-based ketogenic template!

        • “This is why natural seafood has the specific ratio’s it has. We lose that ratio in supplements.”

          This is what makes me wary of supps in general. Eg, whey protein isolate is theoretically a highly bioavailable source of protein, but it lacks all of the other components in dairy. Those components may be necessary to optimally assimilate the protein… and you may be inducing some sort of pseudo-deficiency in the process. “Known unknowns?” perhaps, but the whole food covers all your bases.

          This seems especially true for seafood.

  • Salmon but not fish oil supps improved selenium status:

  • Bioavailability of EPA & DHA measured to be 2-9x higher from salmon than fish oil supps in this study:

    • This study basically confirmed… SIGNIFICANTLY higher bioavailability of EPA and DHA from salmon than cod liver oil:

    • 3.3 grams per day for 2 weeks:

      “Bioavailability of EPA+DHA from re-esterified TGs was superior (124%) compared with natural fish oil, whereas the bioavailability from ethyl esters was inferior (73%). FFA bioavailability (91%) did not differ significantly from natural TGs.”

    • 16 top-selling fish oil products:

      “This study found that over half did not meet their label claims for EPA & DHA, and a quarter exceeded recommended limits for peroxide value.”

  • A fantastic piece Bill. Some critical take away messages here. Will be sharing this with everyone!

  • George

    Does anyone know of an experiment or case study in which a deficiency of an EFA produced atherosclerosis as a symptom?

    • George

      Rephrasing it as the $64,000 question; is the EFA requirement relative to the intake of other fats (meaning you, SFA), or not?

      • rs711

        Yes it is. Membrane composition is a function of those variables (amongst others).

      • you can pry the “American Paradox” from my cold dead hands 😛

        jk, but I do tend to think they’re both protective, thus my cheerleading for seafood & dark chocolate

    • alan2102

      Interesting question. One would have to really dredge through a vast qty of old literature, I think, to get an answer.

      Also interesting are these, tangentially related, to which old Ancel would no doubt have nodded approvingly:
      Circulation. 2014 Oct 7;130(15):1245-53
      Circulating omega-6 polyunsaturated Fatty acids and total and cause-specific mortality: the cardiovascular health study.
      “High circulating linoleic acid, but not other n-6 PUFA [i.e. not arachidonic or DGLA – alan2102] was inversely associated with total and CHD mortality in older adults.”
      Br J Nutr. 2010 Sep;104(6):788-96
      n-6 Fatty acids and cardiovascular health: a review of the evidence for dietary intake recommendations.
      “Data from prospective cohort and interventional studies converge towards a specific protective role of dietary n-6 PUFA intake, in particular linoleic acid, against CVD. n-6 PUFA benefits are even increased when SFA intake is also reduced.”

      and, curiously (please do not mention to ray peat):
      Cardiovasc Res. 2012 Jun 1;94(3):460-8
      Dietary linoleate preserves cardiolipin and attenuates mitochondrial dysfunction in the failing rat heart.
      “The prosurvival effect of the linoleate diet was associated with a greater myocardial content and linoleate-enrichment of cardiolipin, an essential mitochondrial phospholipid known to be deficient in the failing heart. This study demonstrates that, despite having favorable effects on cardiac morphology and function in hypertension, a high-fat diet may accelerate or attenuate mortality in advanced hypertensive heart disease depending on its fatty acid composition.”


      Of course, none of this has anything necessarily to do with “essentiality” or “deficiency”.

  • rs711 “In mammals, this pathway [?4-desaturates 22:5?3 to 22:6?3] has practically been abandoned and an alternate route involving the peroxisome has been generally accepted. This route would imply the microsomal elongation of 22:5?3 to 24:5?3, followed by a second ?6-desaturation step to 24:6?3, which would finally be ?-oxidated in the peroxisome to produce DHA. Again, the same route would work for the ?6 family, finally yielding 22:5?6 from 24:5?6. Whether or not humans have an intrinsic ?4-desaturase remains unknown. Nevertheless, the present paper shows for the first time that the ?4-desaturation pathway is functional in the human species when cells are transfected with the enzyme. We show that when the gene that codes for this enzyme (Fad4) is transfected into human lymphocytes immortalized with the Epstein-Barr virus (EBV), putative ?4-desaturation of both substrates, 22:5?3 and 22:4?6, is produced, yielding DHA and DPA, respectively, as their products.”

    I can see it now – injectable enzymes for the seafood averse/allergic :p

    I love the reverse engineering though, revealing pathways we have apparently abandoned for the most part. Would be interesting to see the artificial ALA/LA-devoid rat diet study you mentioned done in humans (unlucky bastards!) & assay gene expression levels for comparison with other biomarkers

    • the problem with gene therapy, as I see it, is it’s potential irreversibility…

      • rs711

        Yes, that’s certainly an issue. I pointed to that not because of its current viability but because it hints at what our cellular architecture is evolved to handle (more or less) & the legacy of mammalian lipid pathways.

        Regarding gene therapy: the crux of it is the ability to control transcription factors responsible for differentiation (specialization) & dedifferentiation (generalization). So, in principle, if it ‘works’, it’s also reversible & can be modulated. Also, you could ‘just’ transdifferentiate (somatic cell to another somatic cell…not somatic, to totipotent, back to another kind of somatic cell, which is more complicated). Many many options. Many unanswered questions.

        I don’t see any extraordinary inherent danger though. We already know how profoundly our biology can be messed with via much simpler interventions anyways…

  • David Holmes

    What would cause a DGLA deficiency when GLA is normal? Delta-6-Desaturase not working?

    • Does someone have this? It seems very unlikely to have isolated DGLA deficiency… any other lipid abnormalities?

      • David Holmes

        Someone yes, me. ALA < dl, elevated stearic and tricasanoic, low oleic. Rest in range. From Genova NutrEval.

        • Elovl5, but that would predict deficiencies in 20:4n3 (eicosatetraenoic acid) and 22:4n6 (adrenic)…. perhaps a fluke?

          • David Holmes

            2 tests, year and a half apart, same results.

          • Wow! any other theories?

            is it pathological?

          • David Holmes

            6 1/2 years CFS, something is pathological. Fat digestion/absorption is impaired, but not sure there is much DGLA in foods.

    • alan2102

      Delta-6-desaturase is what converts linoleic into GLA, I believe; the desaturation can easily be blocked. GLA to DGLA is supposedly much easier; it is an elongase, not a desaturase.

      If I were you, I would do a full-bore literature sweep to learn all I could about fatty acid chain elongation and influences thereon. You might come up with good potential remedies.

      Here’s a start: pyridoxine stimulates both desaturation and elongation, at least in rainbow trout. Now, I know you’re not a rainbow trout, but still…

      • alan2102

        Also: carnitine is often mentioned in relation to mitochondrial “desaturation-elongation” pathways. I don’t have any more-specific info, i.e. to what extent it actually participates in elongation *per se*. You’d have to drill down into the lit to find out. If you use carnitine, pay careful attention to kinetics. There is recent literature on tissue uptake and retention being stimulated by carbohydrates (like creatine in that respect). It also appears that high single doses are wasted, getting spilled into the urine. Smaller doses, more persistently, seem to be the ticket. Good luck.

      • Given the normal GLA levels, the desaturase isn’t the problem; it *might* be the elongase, but if that was the problem, it should’ve also impacted the n3 pathway… but apparently the n3 pathway was unaffected..?

        • alan2102

          Yes, an elongase, and desaturation is not the issue; see above.

          Should also have impacted the n3s? Maybe. I know nothing at all about elongase specificity (if any), or anything else that might be relevant or that might explain why David has this peculiar problem with GLA->DGLA. Why are his elongases lazy when it comes to GLA, but not to other species? I don’t know, but if I were him I would go ahead and experiment with things that might stimulate elongase activity. I also might try GLA loading, with the idea of “stealing” some elongase activity away from the species being effectively acted upon.

          Hey, David, these might make for useful background reads:

          Regulation of hepatic fatty acid elongase and desaturase expression in diabetes and obesity

          Tissue-specific, nutritional, and developmental regulation of rat fatty acid elongases

          • Tissue-specific, nutritional, and developmental regulation of rat fatty acid elongases


            Of the six fatty acid elongase (Elovl) subtypes expressed in mammals, adult rat liver expresses four subtypes: Elovl-5 > Elovl-1 = Elovl-2 = Elovl-6. Overnight starvation and fish oil-enriched diets repressed hepatic elongase activity in livers of adult male rats. Diet-induced changes in elongase activity correlate with Elovl-5 and Elovl-6 mRNA abundance. Adult rats fed the peroxisome proliferator-activated receptor ? (PPAR?) agonist WY14,643 have increased hepatic elongase activity, Elovl-1, Elovl-5, Elovl-6, ?5, ?6, and ?9 desaturase mRNA abundance, and mead acid (20:3,n-9) content. PPAR? agonists affect both fatty acid elongation and desaturation pathways leading to changes in hepatic lipid composition. Elovl activity is low in fetal liver but increases significantly after birth. Developmental changes in hepatic elongase activity paralleled the postnatal induction of Elovl-5 mRNA and mRNAs encoding the PPAR?-regulated transcripts, ?5 and ?6 desaturase, and cytochrome P450 4A. In contrast, Elovl-6, ?9desaturase, and FAS mRNA abundance paralleled changes in hepatic sterol regulatory element binding protein 1c (SREBP-1c) nuclear content. SREBP-1c is present in fetal liver nuclei, absent from nuclei immediately after birth, and reappears in nuclei at weaning, 21 days postpartum. In conclusion, changes in Elovl-5 expression may account for much of the nutritional and developmental control of fatty acid elongation activity in the rat liver.

          • David Holmes

            Wow, thanks so much guys! I’m afraid some of that is a little over my head 🙂

            n3 looks like it is working well. Here are the fatty acids:

            Also there are many other results on this test that are out of range (amino acids, vitamins, minerals) that might be causative. For example 25(OH)D was 16 ng/mL.

          • alan2102

            “there are many other results on this test that are out of range (amino acids, vitamins, minerals”

            Maybe you intended to post images of your amino/vitamin/etc. results, but posted instead a dupe of your fatty acid stuff.

            Was, by chance, pyridoxine one of the vitamins out of range?

          • David Holmes

            I provided the fatty acids to show than n3 pathway is working well. Pyridoxine was not deficient. Thiamine, riboflavin, niacin, zinc, magnesium, and manganese were.

          • dark chocolate & seafood lover? lots of stearate & DHA 🙂

          • David Holmes

            Oddly enough, a seafood hater all my life until mid 2011. I do eat sushi once or twice a week, salmon and tuna. I’m surprised at the high levels of DHA. Same reading on previous version of this test which was only a year after introducing seafood.

          • alan2102

            Pyridoxine could be effectively in deficit with combined shortfalls of riboflavin and zinc, which influence pyridoxal kinase. Magnesium and niacin (NAD and fambly) are involved in EVERYTHING.

          • Many human vs mouse differences complicate this 🙁

            Eg, elovl2 is a GLA elongase in mice, but doesn’t work so well in humans

            Cool infographic from Linus Pauling Institute:

          • ELOVL2 gene polymorphisms are associated with increases in plasma eicosapentaenoic and docosahexaenoic acid proportions after fish oil supplement.


          • alan2102

            Maybe I missed something, but I did not see in PMID 16564093 (your cite) evidence that GLA elongation does not work well in humans. Maybe it doesn’t, but I could not tell from that abstract.

            It is obviously not working well in David’s case. I wonder if he could modify that?

          • “However, mouse ELOVL2 was able to elongate ?-linolenic acid (C18:3n–6) to some degree, while human ELOVL2 could not, which may indicate a minor functional divergence between the species. “

          • alan2102

            I take it that is from the full text.

          • “It is obviously not working well in David’s case.”

            I don’t know if this is the case, because the analagous fatty acids in the n3 pathway weren’t changed.

            Maybe DGLA is being siphoned into PGs: overactive eicosanoid synthesis could explain why the n3 fatty acids weren’t affected.

          • alan2102

            I see what you’re saying! You are right. I was not accounting for COX/LOX variability, just assuming that they were stable (in which case, sluggish GLA->DGLA was the only visible explanation). No basis for that assumption! So, next question: what if David were to inhibit COX+LOX over a period of time. Would that cause a buildup (plugging the “leakage”) of DGLA over time? And more importantly, would doing so have any favorable effect on him?

  • those who hate eating sea food or fish may get them from supplements. however, I think you can train your brain to like eating a certain way…

    besides, what if we can get to a bargain and use both seafood and supplements? 🙂 just a thought!

  • septeus7

    What about simply ingesting blue-green algae? I’m not referring the Algal-oil capsules but dried algae powder. They sell it in bulk.

    • it should have some DHA, although I don’t know how much you’d need to ingest…

  • Pingback: Weekend Link Love – Edition 318()

  • Pingback: Weekend Link Love – Edition 318 -

  • Pingback: The Roundup()

  • Aaron Ashmann

    Guys, check this thread :

    Also, if you were to consider 2% of a diet of 2000 calories that would be about 4.44 grams of DHA. That seems like a gross overload in terms of DHA over time. Most long lived creatures prefer to keep there tissues saturated with LA and ALA vs DHA and AA.

    • I don’t think that much DHA is harmful if it comes packaged in seafood, where: 1) it’s in it’s most stable form; 2) it’s accompanied by the other micronutrients necessary to properly assimilate it; and 3) vege oils & thus total PUFAs are kept low.

      But I agree that we don’t want overly unsaturated membranes, which is why I recommend saturated fat-rich foods… seafood & dark chocolate FTW 🙂

      • Aaron Ashmann

        I think there are other issues if fat content in the diet gets too high. Check this out:

        Increasing intakes of the long-chain ?-3 docosahexaenoic acid: effects on platelet functions and redox status in healthy men

        Nicolas Guillot


        Docosahexaenoic acid (DHA) can prevent
        cardiovascular events. However, few studies have addressed the effects
        of DHA on both
        platelet reactivity and redox status in healthy
        subjects, and dose-related studies are scarce. The main objectives of
        present study were to determine the effects of
        increasing doses of DHA on platelets and redox status in humans. Twelve
        male volunteers (aged 53–65 yr) were assigned to
        consume an intake of successively 200, 400, 800, and 1600 mg/d DHA, as
        only ?-3 fatty acid, for 2 wk each dose. Blood and
        urine samples were collected before and after each dose of DHA and at 8
        wk after arrest of supplementation. DHA was
        incorporated in a dose-response fashion in platelet phospholipids. After
        with 400 and 800 mg/d DHA, platelet reactivity was
        significantly decreased. Platelet vitamin E concentration increased only
        after 200 mg/d DHA, while p38 MAP kinase
        phosphorylation decreased. Urinary isoprostane was also significantly
        lowered after
        200 mg/d DHA but was increased after 1600 mg/d.
        Therefore, supplementation with only 200 mg/d DHA for 2 wk induced an
        effect. It is concluded that low consumption of DHA
        could be an effective and nonpharmacological way to protect healthy men
        from platelet-related cardiovascular
        events.—Guillot, N., Caillet, E., Laville, M., Calzada, C., Lagarde, M.,
        Véricel, E.
        Increasing intakes of the long-chain ?-3
        docosahexaenoic acid: effects on platelet functions and redox status in
        healthy men.

        I just don’t think even over 1g a day of DHA from fish would be beneficial if your diet is already good.

  • MisterC

    I have eczema and find that GLA from Evening primrose oil (but not borage oil, another source of GLA) improves my skin condition, BUT, it only works when I take it with Omega-3.

    If I take GLA/EPO on its own, no improvement in symptoms and it can make my symptoms worse.
    If I take Omega 3 fish oil on its own, no improvement.
    However GLA + O3 = eczema improves significantly.

    I also have scalp dermatitis and my scalp clears up when I take GLA and O3 together.

    If I take Borage oil, which also contains GLA, no benefit either alone or in combination with O3.

    I know of other people and some studies where people with skin conditions have benefitted from Evening primrose oil, yet the same cannot be said for Borage oil. As both contain GLA which is supposed to be the “active ingredient” could it be that EPO contains something other than GLA that has a positive effect on skin.

  • Jim Jozwiak

    There are other functions of 18:3n-3 besides conversion to EPA and DHA. In mammals, alpha linolenic acid is shunted out to the coat, which in humans means less dry skin, less compacted ear wax, better olefaction.