The day almonds became interesting.

Non-sequiter nutrition: Atwater’s almonds, et al., Op. 87

Almonds have been considered a super-food for as long as I can remember.  And in accord with my level of interest in super-foods, I’ve never cared.  Today, however, almonds became interesting. One small serving of almonds (1 ounce or 28 grams) provide ~171 kilocalories (alternatively, 100 calorie packs have, well, 100 kilocalories).  This measurement of a food’s energy content takes into account the amount of heat produced when it is electrocuted in a bomb calorimeter as well as its digestibility.  The importance of taking both of those things into consideration?  Marshmallows and tree bark produce a lot of heat when they burn.  Unlike marshmallows, however, a tree bark smoothie wouldn’t give us any energy because we can’t digest wood.  This is further complicated because digestibility of a food consumed by itself can differ when it’s eaten with a meal.

Usually, and unlike carbs and protein, the digestibility of fat is impeccably high and unvarying.  Almond oil, however, might be an exception in more ways than one.

Discrepancy between the Atwater factor predicted and empirically measured energy values of almonds in human diets (Novotny et al., 2012)

This was a ridiculously complicated study designed to determine the calories in almonds.  It was a three-way crossover with 18-day feedings of 0, 42, or 84 grams of almonds per day (0, 1.5, or 3 ounces per day).  The researchers gave the volunteers ALL of their food for the entire study, and in exchange, the volunteers gave the researchers the byproducts (urine, feces) for the second half of each feeding period.  This is already an expensive and extremely  labor-intensive study, but I think they were trying to do more than just quantify the calories in almonds; I think they were trying to stick-it-to-the-man.

N.B. the almonds were eaten with normal meals.  The diet was normal.  There are no tricks up my or the researcher’s sleeves.  And I’m honestly fascinated by Table 2.

1.5 servings of almonds (42 grams) had a phenomenal effect on food digestibility.  And 3 servings doubled the amount of non-absorbed calories.  In the beginning of the post I noted that a serving of almonds had 171 kilocalories.  But a serving of almonds increases the non-absorbed kilocalories by about 50.  So does this mean we should re-assign a serving of almonds to 121 kcal?

Yes, the authors decided; and I agree.  And I think this sticks-it-to-the-man.  Perhaps this is the source of almond weight loss lore (?)… imagine the fastidious dieter who weighs out 3 servings of almonds for their daily snack, accounts for the 513 kilocalories in their food diary (but is really only getting 357 kilocalories), and they lose weight…  and those 100 calorie packs only have 68 calories.  Ha!

OK, but just out of curiosity which calories aren’t absorbed?  Are almond calories poorly absorbed, or do almonds block the absorption of other nutrients?

From Table 3, it’s probably fat.  Combined with earlier findings from Ellis, it’s probably almond fat that was trapped inside delicious and crunchy cell walls (Ellis et al., 2004).

In brief, Ellis measured almond fat after three treatments:

1)      Mechanically crushing the almonds

2)      Chewing the almonds (and measuring spit-out almond fat)

3)      Eating the almonds (… and measuring accessible fecal almond fat)

The first two methods didn’t release a lot of almond fat, but the third did, by a little.  As opposed to crushing or chewing, after actual digestion, gut microbes degrade the crunchy cell walls to release the almond fat contained therein.  Unfortunately, however, fat absorption is very inefficient in the large intestine (where this is all happening), which is why the almond fat is either fermented or excreted.

So at this point we’ve got more fat, but also more carbs and fiber from the almond cellular structures making their way into the large intestine (on a high almond diet)… what do the resident microbes have to say about all of this?

A lot, according to a series of studies by Mandalari and his robotic gut simulator  (Mandalari et al., 2008).

Unless you are seriously constipated, the bacterial changes after 24 hours of fermentation are irrelevant.  Looking at 8 hours, which is probably more physiologically relevant, gives us this:FOS, fructooligosaccharides; FG, finely ground almonds; DG, defatted finely ground almonds.

Table 2 (above) is, in a word, perplexing.  Whether or not Mandalari set out to stick-it-to-the-man, he sure did (unless that is just a thing with almonds [?]).  Similar to FOS, almonds had a relatively potent bifidogenic effect.  This is not surprising because of almond’s high fiber content.  What was surprising, however, was that this is completely absent in defatted almonds.  The fiber is the same in almonds and defatted almonds, therefore there is something uniquely magical about almond fat and the long series of unfortunate unlikely events that must occur in order for the bifidogenic effects of almonds to manifest.

The unlikely events: the almond fat must first be protected during chewing and digestion, otherwise it would be absorbed in the small intestine, before it made it all the way to the more “microbial” large intestine.  This is accomplished by almond’s robust cell walls.  Almond fat needs to be released in the large intestine; this requires microbes and is therefore less likely to occur in the small intestine (where microbes are less abundant; if there were more microbes in the small intestine, the almond fat would be released and absorbed before it made it into the large intestine).  The almond fat needs to be not absorbed in the large intestine so it can exert its bifidogenic effect; this happens because the large intestine is inherently poor at fat absorption.  Everything must be exactly in place (kind-of-like in M. Night Shyamalan’s “Signs”): almond’s cellular structure, the intestine’s region-specific digestive enzymes, microbial geography, differential fat absorption capacity, etc., etc.  It’s like an astrological event that occurs once every million years.


Back to the Novotny (Atwater) study for a moment.  48 grams (1.5 servings) of almonds only provide about 5 grams of fiber, but it increased stool weight by almost half.  Fiber is known to increase “regularity,” but the effect of almonds is pharmacologically disproportionate to it’s fiber content.

According to a review by Ahmad (2010), almond oil improves bowel transit time and reduces the symptoms of irritable bowel syndrome.  Not whole almonds.  Not almond fiber.  Almond oil.  And injecting it might even cure IBS (don’t try this at home; Sasaki et al., 2004).

Is it time for a paradigm switch?  Will almond oil open the door for other fats to be researched for bona fide prebiotic properties, akin to inulin and GOS?

Indeed, almonds became interesting today.

For full access to all articles and more (or if you just like what I do and want to support it), head over to Patreon! It’s only five bucks a month and there are many other options.

It’s ad-free and you and you can cancel if it sucks 🙂

Affiliate links: if you’re still looking for a pair of hot blue blockers, Carbonshade is 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 the taste, Real Mushrooms makes great extracts. 10% off with coupon code LAGAKOS.


calories proper



Be Sociable, Share!
  • Savateri6

    Yes!! One of my favorite snacks!

  • Carole

    So interesting!  Any idea how (or whether) the almonds were prepared?  I keep reading about the benefits of soaking almonds (increased digestibility is one — I wonder how that might affect this study), then roasting at low temps (~150 deg)…were these just plain raw almonds?

    • In all three studies, they were California almonds.  The only study to include roasted almonds
      found they were similar to non-roasted. 

      Novotny study: “natural, whole almonds”

      Ellis study: roasted and non-roasted

      Mandalari study: “Blanched finely diced and powdered almonds”
      So basically, yes, they were just plain non-roasted almonds.  I say “non-roasted” instead of “raw” because I think all California almonds are pasteurized.

  • Anna

    So does that mean that reduced calories in almonds would be the same whether they are raw or dry-roasted? Would the dry-roasted Almonds still lead to less calories being absorbed and have a lower actual calorie content? Or does it only apply to raw where they are harder to digest?

    • The almonds used in the study weren’t raw, and other studies showed similar findings from roasted and non-roasted, so it probably applies to all almonds.

  • Bomb calorimetry: 7.11 kcal/g

    Atwater: 6.03 kcal/g

    Empirical data: priceless. 4.6 kcal/g

  • George Henderson

    Good stuff! Ground almonds can be used instead of flour in Ye Olde English Cookinge to thicken Ye Olde English Stews.

  • Pingback: [BLOCKED BY STBV] Almonds are Awesome - Strong Abs()

  • Pingback: People who eat PUFAs, come 'ere. - Page 4 | Mark's Daily Apple Health and Fitness Forum page 4()

  • Pingback: Resistant Starches - Page 126 | Mark's Daily Apple Health and Fitness Forum page 126()

  • Interesting and well written entertainment blog , A+.

    My curiosity is hypothetically many nuts should have the same effect. Two candidates that spring 2 mind are peanuts and pistachios as they are similar fiber rich and moderately high protein nut/legumes.

    Anecdotally (speaking as someone who consumes tons of nuts…):
    -Ketosis is inferior with almonds relative to oily less fibrous nuts.
    -Macadmias and pecans are more ketogenic, due to natural 4:1 ratios
    -Satiety is most superior with almonds and peanuts
    -macadamias and pecans have more pleasant emotional effect, probably secondary to ketosis
    -Though the nutritional composition is similar i have noted marked poor mood when consuming peanuts excessively. Almonds are beneficial or neutral to mood, and pecans/macs tend to enhance mood (likely correlating with ketosis).
    -Almonds seem to potently inhibit appetite although i prefer the way i feel with oily more ketogenic nuts.

    I know peanuts are legumes but SRSLY its like the same thing nutrition wise ;P

    • Thanks!

      Few nuts approach almonds in terms of total fibre (g/kcal) except chestnuts, but they’re super-high carb. After chestnuts is pistachios 🙂

      • Ela

        I love all the detail you give on what’s otherwise floated around almost as a diet urban legend (but one with lots of anecdotal backup and intuitive sense). Can’t find it with a quick search, but sharing Woo’s intuition that this applies to other nuts, and I’m sure I’ve seen an article specifically attributing this property to almonds _and walnuts_.
        Per Woo’s distinction (and perhaps this is also my bias, because these are maybe my favorite nuts): both walnuts and pecans are oilier than almonds (keto range) but still relatively high fiber (yes not close to almonds but way more than cashews, which are too high carb if doing keto).
        Nuts and avocados are among the few foods that have both plentiful fiber and plentiful fat, which (together w/hydration) are the things constantly discussed as promoting good elimination. In my experience, some ppl seem to respond more to one or the other, but handy to have both in the same package!

  • Colin P. Müller

    Excellent writeup. I’ve been thinking about adding more nuts to my diet and removing some animal fats…..though not entirely…I refuse to part with bacon and eggs and fish 🙂

  • Ennsgraber Christian

    What do you think about coconut flour / flakes / meat? Are you aware of any studies which show a similar mechanism of feeding gut bacteria via fat in the cell walls?