From Neroli to NuGrape: Methyl Anthranilate

Oof! It's been a while since I've posted anything here. My excuse is that I've been writing, or pantomiming writing, or sitting in front of my laptop furrowing my brow and wondering, "what is it... to write?" I think this is a pretty common dissertation symptom. Writing ceases to be a series of deliberate actions and instead becomes a sort of misty tunnel that you enter and exit each day wondering, "What happened? What is happening? Is this real life?" But! I have a couple of other blog posts on the transom, "somewhat finished," and so I promise that there will be new material here more than semi-seasonally.

In the meantime, here's a preview of something that I might talk about next week at my Fellow in Focus lecture here at Chemical Heritage Foundation. (The lecture is free! So if you're in Philadelphia on April 2, come out and hear me talk about this stuff in real life!)


The question I'm starting from is this: if you wanted to make a flavor additive, in or around 1920, what would it take? What would you need to know? What would you need to have access to?

The first thing to realize is the most obvious. Making synthetic flavors meant working with what was available -- in terms of both knowledge and materials.

When it came to knowledge -- that is, certain knowledge of the flavor chemicals actually present in foods -- for much of the first half of the twentieth century, there was little to go on. Even as other material components of foods -- proteins, carbohydrates, fats, vitamins -- were chemically determined and quantified, flavor research lagged behind. There are several reasons for this. Usually, flavor chemicals are only present in tiny amounts in food -- parts per million or even less. In early twentieth-century chemistry laboratories, isolating and identifying chemicals present in such small quantities was tricky, and labor- and material-intensive. (For instance, USDA chemists in the early 1920s attempting to identify the chemicals that gave apples their aroma had to start out with nearly a ton of apples to get less than two grams of aromatic material for analysis). Complicating matters further, flavor chemicals are often volatile, unstable, and reactive. It took meticulous work to ensure that the chemicals identified in the final result were not artifacts created in the process of analysis. Which is all to say that identifying the chemicals responsible for flavor in foods is a very difficult problem, and, until the 1950s -- when powerful analytic technologies such as gas chromatography became available -- very few people attempted it.

E.J. Kessler's Practical Flavoring Extract Maker from 1912.

E.J. Kessler's Practical Flavoring Extract Maker from 1912.

So, in most cases, when a maker of flavoring additives circa 1920 was formulating an artificial "strawberry" or "pineapple" flavor, he (almost always he) was not pretending to reproduce the natural world on a molecular level. That is, he was not trying to synthetically replicate the actual chemical components of actual pineapples. He was working from standard chemical recipes gleaned from formularies, handbooks, or trade journals, or kept under lock and key as company secrets. He was also using his sensory and scientific knowledge of different chemicals, so that he could combine available materials in appropriate ways to obtain desired qualities (a "fresher" tasting peach, a strawberry flavor that was suitable for candy lozenges.)

Getting the raw materials for flavor-making meant shopping in the same chemical marketplace as perfumers, pharmacists, and soap and cosmetics makers. Supply houses such as Schimmel & Co., W.J. Bush & Co., Synfleur, and others typically sold both proprietary perfume and flavoring formulations and "raw materials" for the industry -- synthetic aromatic chemicals or purified isolates, natural essential oils, extracts and essences. Frequently, the same chemical would be put to work in different contexts, appearing in different types of products, producing distinct effects, acquiring different meanings.     

Which brings me to the story of exemplary chemical: methyl anthranilate.

By the turn of the twentieth century, methyl anthranilate was already an important chemical for perfumers. In the mid-1890s, it had been identified as a key component of neroli -- the essential oil of orange blossoms. Its presence was subsequently discovered in other natural essences: tuberose, jasmine, gardenia, ylang-ylang, and bergamot. In other words, methyl anthranilate was a frequent chemical denizen of the lush pleasure gardens of early twentieth-century floral perfumes, scenting a lady's handkerchief, or the bosom she held it to.    

I mentioned earlier how tough analytic organic chemistry could be? People in the essential oil and perfumery business needed to be well-versed in its techniques and methods, and to have a comprehensive analytical understanding of the chemical components of their materials. Essential oils are costly; they vary in quality; dealers can be unscrupulous. Careful chemical analyses could not only detect frauds, but also determine purity, and thus value. Knowing the chemical components and physical properties of essential oils was necessary to staying in the business.

An advertisement from 1899 for Schimmel's Synthetic Oil of Orange Blossoms, "identical with the oil distilled from Orange Flowers." Methyl anthranilate was a crucial component in this compound.

An advertisement from 1899 for Schimmel's Synthetic Oil of Orange Blossoms, "identical with the oil distilled from Orange Flowers." Methyl anthranilate was a crucial component in this compound.

Some, however, turned their analytic knowledge of the chemical constituents of essential oils to commercial use, by manufacturing synthetic versions of chemicals present in natural oils. This is how synthetic methyl anthranilate began to be produced and sold, as "artificial neroli oil." I'm still trying to figure out exactly how methyl anthranilate was manufactured synthetically, but according to an 1897 article in the Journal of the Society of the Chemical Industry, one way was to combine methyl alcohol with anthranilic acid under an inverted condenser, and then saturate it with gaseous hydrochloric acid.

In any case, in the first decades of the twentieth century, methyl anthranilate was sold by major perfume material supply houses such as Schimmel, Van Dyk & Co., W.J. Bush & Co., alongside both "synthetic" essential oil blends and natural materials.   

 But methyl anthranilate doesn't just smell like springtime and orange blossoms and fancy, old-fashioned ladies. Diluted, it has a distinct quality that many of us would find familiar: the odor of grape jolly ranchers, or grape soda, or any of the deep purple sweets of indiscriminate childhood.

The affiliation of methyl anthranilate with grape-flavored soda and candy dates back to the beginning of the twentieth century, when it became a widely available chemical material. People who worked with flavors began using methyl anthranilate in flavoring syrups used for grape soda pop, candy lozenges, and other grape-flavored things. They also used the chemical in in other fruit flavorings: banana, orange, and pineapple.

Let me underscore one point: when perfumers first used methyl anthranilate in their synthetic perfumes, they knew that the chemical could be found in actual neroli, jasmine, and so on. When flavoring manufacturers first adopted it for use in their fruit flavors, they had no way to make the claim that the chemical was an actual aspect of the "true fruits."

But, in addition to essential oil dealers, there was another group of chemists who were interested in analyzing and cataloguing the chemical contents of natural materials: government regulators at the USDA Bureau of Chemistry and in state health agencies, who were responsible for enforcing the 1906 Pure Food and Drug Act. In addition to monitoring the safety of the food supply, the law also aimed to protect consumers against fraud -- to protect them from being deceived by sophisticated chemical additives into taking "imitation" goods for the real thing. The law created a statutory distinction between "natural" and "artificial" in the food system. Foods that included synthetic flavor additives would have to bear on their labels the scarlet letter that declared their second-class status: ARTIFICIAL.

According to the law, the unannounced addition of synthetic chemicals like methyl anthranilate to soft drinks, jams, and so on constituted illegal adulteration. Violators faced a seizure of their goods, fines, and subsequent loss of business. But to enforce the law, regulators had to prove that the food in question contained a chemical additive.     

And this proved to be a problem. As the Journal of the Franklin Institute put it in 1922: "Inasmuch as methyl anthranilate in a dilute form possesses a decided grape-like odor, its detection in commercial grape juice appears to have led to the conclusion on the part of some of those engaged in the control of these products that in all cases of its occurrence an artificial flavoring agent has been employed."

But in fact, this was the wrong conclusion to draw. As researchers at the Bureau of Chemistry discovered while trying to develop official methods for proving that synthetic methyl anthranilate had been added to foods, the chemical was present not only in artificial grape flavoring, but also in actual grapes. Frederick B. Power, the head of the Bureau's phytochemical laboratory, and his lab partner Victor Chesnut, did not find it in Vitis vinifera grapes, the "old world" European varietals. But they did find it in the foxy, foxy Vitis labrusca and other grape varietals of the New World: Niagara, Catawba, Delaware grapes. Concord grape juice, in fact, contained the highest concentration of the chemical. So, in trying to find a way to determine the presence of a chemical adulterant, Power and Chesnut confirmed the chemical's presence in actual grapes.

So far, we've followed methyl anthranilate from its identification in "natural" Neroli oil, to its synthesis for use in synthetic perfumes meant to imitate this sensation, to its inclusion in artificial grape flavors, to the discovery -- by government regulators -- of its presence in actual grape juice.  

Part of what this story should suggest is the problematic distinction between "natural" and "artificial." Molecules like methyl anthranilate are discoverable in haunts throughout the natural and artefactual worlds, appearing in various guises, for various purposes. At different concentrations, in different contexts, they have different effects and properties. For instance, one of the current uses of methyl anthranilate is as a bird repellent. Asking whether something is "real" or "fake" tells you less about the thing in question, more about the social and cultural contexts in which that thing is evaluated and exchanged.  

(This is also, by the way, one of the reasons it's ridiculous to claim that a chemical shouldn't be in foods because it's also in yoga mats, or whatever. Its presence in both the edible and non-edible world has absolutely nothing to do with whether it's toxic, or good, or gross, or anything.)

My chemists -- the ones who prance through the pages of my dissertation -- will most likely tell you that a molecule is a molecule, that it's impossible to distinguish a molecule of methyl anthranilate within a Concord grape's glaucous globe from one produced in a laboratory by mixing chemicals under a condenser hood in the presence of hydrochloric acid gas.

But I'm not a chemist; I'm a historian. And even if there is no distinguishable chemical difference between two molecules -- one synthetic, one "natural" -- there are historical differences, and those differences have a meaning. Things have histories, things come from somewhere, and how they got here matters. Tracing the history of flavors means following the threads of all these material and sensory entanglements -- chemicals, workers, technologies, laws, markets, foods, consumers... 

Some people reading this might know that the origin of this whole research project started with grapes, or maybe with methyl anthranilate. The short version: once, I was tempted to try a dusky violet Concord grape at the Union Square farmers market. "Wow," I thought. "This totally tastes like fake grape." I wondered whether the Concord grape was more common back when "fake grape" was "invented."  "Maybe 'fake grape' was supposed to taste like real grapes, only these were the real grapes, back then." 

I've spent the past two years and change on the trail of this idea, mostly learning how to ask the right questions.      

On a final note, here's the excellent NuGrape song, recorded by the mysterious and beuatiful "NuGrape Twins" in 1926. I first heard it on the collection American Primitive, Vol. II, on Revenant Records, but you can listen to it here.

This is how it begins (lyrics transcribed by Michael Leddy):

I got a NuGrape mighty fine
Three rings around the bottle is a-genuine
I've got your ice cold NuGrape
I got a NuGrape mighty fine
Got plenty imitation but they none like mine
I got your ice-cold NuGrape...