Nose Dive by Harold McGee
This book is about Harold McGee (author of On Food and Cooking) going on a ten year journey through the world of smells. It was instigated by the first time he had aged grouse and found the smells at first virulently repellent, but later pleasant, and extremely interesting.
We have 400 or so different types of odor receptors, packed into an area of about a square inch.
In the lab, gas chromatography-olfactometry is used to isolate and describe smells. GC is used to separate the components of a complex mixture, and as each component comes out, a human smeller sniffs and identifies it, and thus the whole complex mixture is characterized into smell components.
He takes an interesting approach to exploring smells, and begins in interstellar space, where the smells would be mostly unpleasant to us. They’re either “chemical” because the molecules there are the same as the materials that we use on Earth to burn or clean or fumigate, or they’re disgusting because they’re prominent on earth in decomposing plant or animal remains. All these molecules lack oxygen. Once oxygen is added to carbon chains, it begins to smell more like life. Most molecules from animals and plants contain oxygen, along with carbon and hydrogen.
He introduces the starter set of life’s volatiles: a bunch of lightweight, small, simple carbon chains. They fall into four main chemical families - the acids, alcohols, aldehydes, and hydrocarbons - and the smells within each family often bear a resemblance.
The medium-length carbon chains, by contrast, are more likely to be volatiles purposely generated by plants, animals, and microbes as signals: the smells of flowers and fruits that are meant to catch the attention of insects and animals, often to attract them. The smells of purposeful signals from animals is much more diverse than the smell of animal bodies tends to be.
Hydrocarbons as having chemical smells because we encounter them in prominent amounts only in fuels and solvents. Nevertheless, traces of some hydrocarbons are present in many foods and in our own bodies, and contribute to life’s olfactory hum. So McGee often describes smells from certain fruits or other plants as “solventy.”
The smells of our own bodies can be found in the plant and animal kingdom. McGee describes the basic recipe for armpit smell as “touches of vinegar, cheese, metal, mushrooms, rancid goat, meat, grapefruit, and onions.” McGee raises the interesting idea that for early humans, perhaps encountering whiffs of themselves in certain strange new foods made those initially odd materials more approachable to eat. For example, cheese is a solid concentrate of proteins and fats, the same major ingredients of our body, and the surface of freshly made cheese is readily colonized by some of the same protein- and fat-eating bacteria that live on our skin. In modern day, McGee says we all find the weird smells of our bodies delightful, even if we don’t admit it. So foods that have some of those same smells allow us to have those delight experiences, but at a remove, and at times of our choosing, thus skirting the social embarrassment.
Some smells aren’t exactly pleasant to us, like the smell of a strawberry or rose, and many people actively dislike them. But they catch our attention, momentarily make us “all nose”, because they were once central to our animal life, and fleetingly restore that lost sensory dimension to it.
Humans tend to find most plant volatiles pleasant, as opposed to animal volatiles. Instead of those mostly small, incidental, funky waste products of metabolism from animals, plant volatiles are generally larger molecules, six to fifteen carbon atoms long, actively built by the most prolific synthetic chemists on the planet. Plant smells mostly consist of oxygen, carbon and hydrogen. With animal smells, nitrogen and sulfur enter the mix. Animals contain about twice or more the amount of protein per weight than plants.
One special type of smells from plants is esters. Esters are hybrid molecules that result when one acid molecule and one alcohol molecule are jammed together at their oxygen-bonded ends, knocking out a molecule of water in the process. Fruits are the plant kingdom’s main ester-making factories. They ramp up production to signal their ripeness when their seeds are mature. Most fruits make a blend of a dozen esters or so. This means that different fruits emit many of the same esters, but in varying proportions that give them their own overall smells. The biological function of esters remains enigmatic. They may be a by-product of metabolism, or a way of making both acids and alcohols less toxic to the cells that produce them, or programmed signals for attracting mobile insects - any or all of the above.
Another class of molecules plants secrete is terpenoids. Volatile terpenoids may have been initially developed for signaling by early land insects, then adopted as insect-confusing defenses by early land plants. It’s shared terpenoids that make ginger and lemongrass resemble Amazonian ants. This pattern of land insects using a molecule for signaling, only for it to be synthesized by plants to confuse and manipulate them is a common pattern we see over and over again. Terpenoids are found prominently in pine trees and the peels of citrus fruit.
Plant molecules that smell pleasant to us, like generically flowery linalool, are actually defensive molecules. Fruits often amp up production in response to attacks by mold, and it harms the mold when linalool levels increase.
Benzene rings are very important to the evolution of plants. Their ability to absorb UV rays may have allowed early algae to move into land plants over time. Benzoid molecules from plants, like vanillin in vanilla, often smell nice to us, but are usually defensive molecules. They create the world of spices!
Interestingly, pigment and scent volatiles often share resources, so the more colorful a plant is, often the less it smells.
The volatiles emitted by flowers often attract one species, but repel others. Many of these smells are not “flowery” to us, and may be similar to the smells emitted by the very first flowers, which evolved to attract flies, which are the most ancient pollinators. Flowers often emit different sets of volatiles in the day vs night, depending on what they’re trying to attract.
Scented flowers that appeal to a wide range of pollinators tend to produce a mix of simple-chain, terpenoid, and benzenoid volatiles, in effect covering all the bases, although one category usually dominates the other two somewhat. Honeybees seem to have a broad palate. They show some preference for terpenoids but are also attracted by spicy and fruity benzenoids and the complex mixtures in roses. Butterfly flowers tend to be brightly colored and to have a mild, fresh, sweet scent coming mainly from a couple of floral-honey benzenoids, phenylethanol and phenylacetaldehyde, and the common floral terpenoid linalool along with fresh-woody ocimene. Moth flowers form a distinctive group and include many of the flowers we think of as exotic and intoxicating.
Interestingly, sometimes an unpleasant element in a complex odor mixture promotes “attentional capture” - gets the brain to commit more resources to processing the sensation - and this strengthens and prolongs the overall sensation. Indole is an example, it usually signals animals and decay, both potential threats, but is found in smells we typically find pleasant.
The English essayist Walter Pater wrote that “all art aspires to the condition of music,” to music’s emotional immediacy. I’ve long thought that all cooking aspires to the condition of ripe fruit.
In the oceans, most of the wild plant biomass produced every year is consumed by animals, but on land, less than a fifth is eaten. In the oceans, animals are estimated to outweigh plants by a factor of thirty; on land, plants outweigh animals by a factor of a thousand. These things contribute to the drastically different smells of these two environments. Dimethyl sulfide is the typical smell of the sea, it is the most abundant molecule made by aquatic creatures, mostly by plankton. DMS is a general signal for the entire oceanic food web that there’s food available. Animals at all levels, from krill to filter-feeding whale sharks to seals, are attracted to DMS-emitting communities of plankton. Seabirds often zero in on prey in their thousands of miles flights over the open ocean by detecting plumes of DMS. Unfortunately, plastics tend to accumulate DMS, which may be one reason they’re often confused with food by ocean animals.
The smells of intact fish and shellfish and seaweeds are often leafy, mushroomy, and cucumber. But once the energy supply runs out and the living systems fail, the production of these chains slows and finally stops. Those that escape into the air aren’t replaced, and instead, other volatiles develop and accumulate. Oxygen in the air takes over the job of chopping up the long carbon chains, especially at their vulnerable kinks. This nonspecific, nonenzymatic breakage produces a different spectrum of fragments, including seven- and ten-carbon chains that are seldom found in living tissues.
230 molecules account for the majority of important volatiles in a wide range of foods, and a given food aroma can be reasonably simulated with around a dozen of these key molecules in specific proportions.
Discussion questions:
To what degree is our sense of smell a reflection of what was evolutionarily important to our ancestors rather than what is relevant to survival today?
It would be interesting to know when smell arose evolutionarily and from what. It seems like aquatic animals would need to develop something more like taste than smell because of the medium.
How much about our dislike of certain smells is cultural or based on memory and context? Apparently babies don't mind the smell of excrement, even though his sort of contradicts the notion that we are naturally repulsed by certain smells (like sulfur). We learn to enjoy smells like butyric acid, which appears in both cheese and vomit. To what degree can you train yourself to approach smells with interest instead of disgust? How much of our disgust in smells necessarily relates to how we first encountered the smell?
We were struck that most of the cited research on smell typically used "untrained" smellers. It kept making us curious how far you could take your smelling ability. Can humans identify particular types of illness like dogs if we train hard enough? Can we track down scents? How long would it take to become an excellent smeller?