In his September wrap-up of his Northern New Mexico Loop trip Paul “PMags” Magnanti mentioned his love for cotton-polyester blend thrift store hiking shirts, particularly western-style snap-button shirts. Having been raised in a “cotton kills” household I found this to be rather shocking, but since he obviously does a lot more hiking than I do I decided to do a little more digging.

I picked up a Kuhl Airspeed (265g, $85) and Wrangler Riata (194g, $20) to compare against my Montbell Wickron Light (203g, $69). The Montbell is 100% polyester and the Wrangler is 55% cotton and 45% polyester. The Kuhl is a bit more complicated since it’s composed of three different fabrics: a 72% nylon 28% polyester blend, an 85% polyester 15% cotton blend, and an 85% polyester 15% cotton mesh. Since the most cotton in any of those fabrics is 15% the upper bound on the amount of cotton in the Kuhl is 40 grams whereas the Wrangler has around 106 grams. If cotton is really as bad as everyone says it is then I’d expect the Wrangler to perform much worse than the Kuhl which should again be noticeably worse than the Montbell.

For the actual experiment I filled up a sink full of cold tap water and dunked each of the shirts in it, kneading them to ensure they were 100% saturated. This was intended to simulate a worst-case scenario where you’re completely soaked either by falling into a body of water or from prolonged rain. I then weighed each dripping-wet shirt and placed in on a hanger on my shower curtain rod. The ambient temperature in the cabin was 67 degrees per the thermostat and I turned the bathroom fan on to introduce air movement as well as to prevent a buildup of humidity. In the first run I recorded the weight of each shirt in 30 minute intervals until the shirts were dry. However, the drying rate for the first 30 minutes is significantly faster so I performed a second run where I recorded the weights every five minutes for the first half hour and every 30 minutes after that.

Here’s a link to the raw results and charts:

https://docs.google.com/spreadsheets/d/1Xuwc2rqYqBKl_bcQo9W1R-1eBS9Dg9LqAE84SR54xgI/edit?usp=sharing

Water Held

The first thing you hear about cotton is that it holds more water than synthetics (I’m using “hold” and avoiding the terms absorb and adsorb since those have very specific technical meanings which we’ll get to later). This graph shows how much water was held by each shirt over time:

We can see that despite being 55% cotton the Wrangler and the Montbell are neck-and-neck as far as the amount of water held whereas the Kuhl has significantly more. Since all the shirts are the same style and size the only other variable is their initial dry weight, so let’s see if there’s a relationship there:

Aha! We can see that there is a linear relationship between the dry weight of a garment and the amount of water that it can hold. The high coefficients of determination (0.935 and 0.901) indicate that this relationship accounts for 90% of the variation in amount of water held, meaning that fabric choice accounts for less than 10%. Of course, three data points does not a study make but the literature backs this up as far as fabric thickness (I’m using weight as a proxy here since the shirts are all the same size and style).

Finally, let’s try another visualization that compares water weight as a percentage of dry weight:

Recall that even though the Wrangler is slightly lighter than the Montbell it’s actually holding relatively more water. This may be our first indication that cotton is making a difference. Let’s do some math based on the 25th minute’s weights (a relatively stable data point in drying rates once all the initial drip-drying was complete):

  • The 218g Montbell was holding 364g of water. Since it’s composed entirely of polyester we can estimate that one gram of polyester holds 1.67 grams of water.
  • The 193g Wrangler is 45% polyester so it contains 87g of polyester and 106g of cotton.
  • Those 87g of polyester should hold 145g of water per our prior calculation.
  • Since the Wrangler held 354g in water at that time, the remaining 209g must have been held by the 106g of cotton.
  • Therefore one gram of cotton holds about two grams of water, about 20% more than polyester.

So wait, what about all this “cotton absorbs 25 times its weight in water” nonsense? Even accounting for differences in polyester varieties between the Montbell and Wrangler this obviously cannot be true. The problem is that people confuse the properties of cotton the fiber with cotton the fabric. The vast majority of water held by fabrics isn’t inside the individual fibers (absorption-with-a-B), it’s trapped in the spaces between the fibers in the fabric (adsorption-with-a-D). So even if an individual cotton fiber can hold 25 times its weight in water this is completely dwarfed by the capacity of the fabric itself. Intuitively think about how much “space” is inside a teeny tiny fiber vs. between strands and layers of fibers in a fabric. In other words, the absorption capabilities of the individual fibers is irrelevant.

Conclusion: the amount of water held by a garment is determined by weight and not the choice of fabric.

So why the bad rap for cotton? Since nylon and polyester fibers are much stronger than cotton my intuition is that synthetic garments can me made significantly lighter (thinner) than their cotton counterparts while maintaining similar levels of durability. In other words cotton clothing has historically been heavier and thus holds more water. However as we can see with the Wrangler this weakness can be remedied by blending it with a synthetic.

Drying Rates

Another thing we often hear about cotton is that it dries more slowly than synthetics. If this were true then we’d expect that the Montbell would dry the fastest followed by the Kuhl and then the Wrangler should be lagging behind (recall that they have 0%, up to 15%, and 55% cotton respectively). In fact, since the Montbell and Wrangler are also comparable in terms of weight we would expect those two shirts to be an apples-to-apples comparison exhibiting noticeably different drying rates.

Here’s a chart comparing the drying rates of the three shirts:

Once drip-drying ceases around the 20-minute mark the rates stabilize and all the shirts are identical up until around 120 minutes. At this point the Kuhl maintains a constant drying rate for the next two hours while the Montbell and Wrangler continue to taper off.

Once again it’s obvious that the choice of fabric does not influence drying rate since the Montbell and Wrangler are indistinguishable. And although the Kuhl is heavier it meets or slightly exceeds the drying rate of the other two shirts across the board, so it appears that weight is also not a significant factor.

During testing I noticed that drying happens “top-down” in the sense that the first places to dry on the shirts were the tops of the shoulders. The collar would remain damp longer but it’s also typically thicker than other parts of a shirt and thus holds more water. After that point the dryness would continue downwards until the only parts of the shirt that remained damp were the cuffs which are also thicker material.

My suspicion is that drying rate is proportional to surface area and that it is also uniform across the surface. Intuitively this makes sense because evaporation happens at the boundary between the fabric and the air so the larger the boundary the more evaporation possible. The wet collar can be explained by weight: there’s more material and thus more held water so it will take longer to dry. The wet cuffs can be explained by gravity: as moisture leaves the garment it “frees up” water holding capacity and gravity then pulls water down from above. So even though the garment is not dripping it behaves almost as if the water is “draining” out through the bottom. Finally, I believe the Kuhl’s seemingly elevated performance can be explained by its higher weight: as a garment dries top-down there is less wet surface area and thus the drying rate tapers off. Since the Kuhl’s heavier fabrics hold so much more water its surface remains fully wet for a longer period of time compared to the other two shirts.

Absolute drying rate is all fine and dandy but recall that each shirt holds a different amount of water and thus has a different amount of drying to do. If all three shirts have equal drying rates then whichever holds the most water will have the longest drying time. So, let’s try converting our drying rates to be relative to the initial amount of held water:

Once again the Montbell and Wrangler are tied but here we see that the Kuhl initially dries at a relative rate that’s about 75% of the other two. This is because although they have similar surface areas (and thus drying rates) the Kuhl holds significantly more moisture due to its higher weight.

Insulation

Drying times aside, the last thing we hear about cotton is that it loses its insulation value when wet. This is true, but it’s also true for fabrics in general. According to a more recent study (pdf), as a garment becomes saturated thermal conductivity (the inverse of insulation) increases linearly from the fabric’s dry thermal conductivity to that of water. Past the saturation point the conductivity is identical to that of water so you might as well just be naked and wet. If you want to stay warm, stay dry. Other studies show that absorbent materials such as wool and cotton can hold a fraction of their weight in water without any change in conductivity. This explains the common adage that wool can get wet without feeling wet and also explains why natural fabrics often feel more comfortable against the skin than synthetics.

If you like anecdotes (and I know I do) I’ve done two fall day hikes using the Wrangler, one in 34F weather and one in five hours of rain, and I certainly didn’t feel any colder than either the Kuhl or Montbell despite being soaking wet and in lower temperatures than the rest of the 2019 season. One thing worth noting is that if your clothing is loose (as the Wrangler is since it’s a cowboy shirt) then it makes very little contact with your skin which will greatly reduce the amount of conductive cooling.

Conclusion

I love my Montbell for its practicality and good looks, it works as both a hiking shirt and around-town shirt. The Kuhl is interesting – the mesh panels are incredibly breathable but the heavier fabric (I assume intended for bushwhacking) in the rest of the shirt is not. It also doesn’t come in plaid. But I have to say I’m with Paul on this one: for a measly $20 you can get a good-looking hiking shirt that’s lighter, more breathable, and dries faster (or just as fast) as the Kuhl or Montbell at a quarter of the price.

Oh, and it won’t kill you.

Caveats

The experiment is only useful for comparing drying times of the shirts in those specific conditions and is best suited for a relative ranking of performance. Obviously your body heat, movement while hiking, and weather conditions will massively influence real-world drying times.

Further Reading

History

  • 2019-12-09 Initial version
  • 2019-12-18 I wasn’t happy with the Insulation section so I did some more reading, bought some studies (at $40 a pop, yikes) and added a paragraph summarizing the results