Posted on

Herder ating marijuana seeds

Higher Profile: Doug Fine, Hemp Farmer and Goat Herder

Doug Fine is known for evangelizing goats and gardens to save the planet, as well as advocating for the power of hemp.

Doug Fine with Deborah Colburn, Interpretive Program Supervisor, harvesting hemp manually for the first time in 100 years at George Washington’s farm at Mount Vernon, in Fredericksburg, Virginia. Photo Courtesy of Doug Fine.

Doug Fine wants to save the planet by teaching humans about a regenerative and sustainable lifestyle. A lofty goal for a hemp farmer and solar-powered goat herder, but Fine persists. That’s the thing about saving the planet, it takes tenacity. It takes Evangelizing in the Biblical sense, from our mouths to their ears. They may not want or be able to walk the talk, but they will hear you.

Author of six books to date, Fine’s first effort, Not Really an Alaskan Mountain Man, was published in 2004, reflecting his introduction to nature as a guy who grew up in the suburbs of New York. Another published in 2008, Farewell My Subaru, details his life living “green off the grid,” demonstrating how to drastically reduce the use of fossil fuel in order to live sustainably. This was followed in 2012 by Too High To Fail, with a focus on the regenerative side of the emerging cannabis industry at the time, and the green economic revolution—that’s now in full swing ten years later.

In 2013, he appeared on TEDx Talks in Albuquerque, New Mexico, where his farm, the Funky Butte Ranch, is located in a remote area hours from the nearest city. The talk, tilted “Why we need goat herding in the digital age,” is a call to arms, with the intent of luring humans back to the garden to save their soul—and health.

Fine introduced himself: “I stand before you today, a neo-rugged individualist, solar-powered goat herder.” Thus begins his humorous-yet-informative talk on how and why he supports his family by tending goats off the grid.

In 2014, he published Hemp Bound: Dispatches from the Front Lines of the Next Agricultural Revolution, wherein he shares his life on his farm, expounding on the many uses of hemp—and how it can help save the planet.

His latest effort was published in 2020, American Hemp Farmer, Adventures and Misadventures of the Cannabis Trade, wherein David Bronner, CEO of Dr. Bronner’s Magic Soap, exuded, “A fantastic piece of Americana that shows the way to a sustainable future.”

American Hemp Farmer has been developed into a TV series, with a pilot and episodes in the can, and more in production now, seeking distribution.

The series includes visits to the Rosebud Sioux tribal lands, with Fine advising on its organic hemp cultivation. Other visits within the show include George Washington’s Mount Vernon estate, with Fine manually harvesting, wearing a full outfit of Colonial-style clothes made of hemp, of course.

Doug Fine: Evangelizing a Sustainable LIfe

As detailed within his TEDx Talk, his first experience with nature was moving to rural Alaska in 2003, where learned about sustenance fishing, catching salmon in the wild.

He enjoyed the thought of sourcing food from the backyard, so to speak. This, he said, got him in touch with what he calls the Indigenous Gene, or I-gene, calling humans back to our Paleolithic roots from living off the land as hunters and gatherers.

“Despite all our digital age accouterments, as humans, we are still the same hunter-gathers that we’ve been for tens of thousands of years,” he said. “I feel at my absolute best self and more relaxed when I’m out milking a goat at first light of day, with the local owls returning from date night. For me, it’s this feeling of living as one is intended to live.”

The experience in Alaska reawakened a vital part of himself that he’s been cultivating ever since, moving to New Mexico two years later, establishing his Funky Butte Ranch, to nourish his soul, with the end result of giving him a sense of contentment. Balance, he said, between the digital age and our indigineous selves.

And then there’s Climate Change, for those who understand the ramifications.

“We’re at the bottom of the ninth with two outs when it comes to tackling climate change, and we’ve got a game plan,” he advised. “Teaching that to everyone is my day job.”

And teach he does, with courses offered from his website, as well as hundreds of speaking engagements around the world under his belt.

To date, the most high profile talk given was a plea to the United Nations, in association with The European Coalition for Just and Effective Drug Policies (ENCOD), an organization working for better drug policies, globally. In this nearly five minute talk he urged change within the failed War on Drugs.

On February 27, he’ll be the Keynote speaker at SXSW’s Eco-Ag Conference in Montana for its 50th Anniversary, with the event airing on C-SPAN.

His “Johnny Hempseed” journey teaching the citizens of Earth how to help heal the planet is seemingly endless, as he presents himself clad head to toe in hemp—including hemp boxers made by his longtime companion.

Doug Fine harvesting hemp with a scythe at George Washington’s Mount Vernon Estate. Courtesy of Doug Fine.

Hemp can Heal the Planet

The stats on how sustainable industrial hemp is are remarkable, when one thinks of all the trees felled over the years—not to mention the amount of plastics now littering the earth that could have been made with hemp and other plants.

“The prohibition of cannabis, and subsequently industrial hemp, was a terrible mistake that a great country made,” he explained from the ranch. “In my talks, I bring with me a little plastic goat made of hemp, created using a 3-D printer. We don’t need to use petroleum byproducts—we never did.”

The benefits of industrial hemp are many, able to be used for everything from fuel to building materials, to pulling toxins from the ground after contamination—demonstrated at what is now Ukraine, at the site of the Chernobyl nuclear melt-down, where thousands of hemp plants have been planted.

Fine’s own hemp seeds from his farm are being used in an experiment to clean contaminated soil in a New Mexico University study, with initial reports of great success in pulling uranium.

“I can confidently write that hemp cleans up radioactive soil,” he wrote within a blog at Vote Hemp. “Not, I heard it does, or I wish it did, or even someone told me they used it at Chernobyl. It actually does, according to this study.”

As explained in an article published by the Global Hemp Associations, the process is called Phytotech, wherein plants can actually decontaminate soil by pulling toxins—with hemp being exceptionally good at the process, decontaminating at a very high rate, eating up chromium, lead, copper, nickel, and more.

Cleaning air quality and soil is nothing new for plants, but our understanding of how they work is.

“When you look at how many trees it takes to make anything, and how many years it took for those trees to grow big enough to use, it’s stunningly ignorant of us to ignore these facts,” he explained. “Before we began synthesizing petroleum byproducts, everything we made and used came from the earth—and it was all regenerative and sustainable. There’s absolutely no reason why we can’t turn this around.”

To give one example, as noted by the European Industrial Hemp Association, hemp contains upwards of 65 to 70 percent cellulose, whereas wood measures in at around 40 percent. The Ministry of Hemp informs that one acre of hemp can produce as much paper as four to 10 acres of trees over a 20 year cycle. Hemp stalks grow in four months, whereas trees take 20 to 80 years, depending on the species.

One can see why the “Plant for the Planet” movement was founded, encouraging humans to plant as many trees as they can—with the goal of one trillion trees planted globally by 2030.

“It’s such a no-brainer,” Fine lamented. “Hemp paper is more durable than paper made from trees, because it doesn’t break down over time. Building materials made from hemp are also mold and fire resistant. Not to mention the devastating effect deforestation has on the climate and health of the planet.”

Climate Change at the Door

Several years ago, a massive, 130,000-acre wildfire hit the Funky Butte Ranch, devastating years of hard work on the farm.

“This is not a dress rehearsal, it’s really happening now, and it’s at the door” Fine said of climate change and the forever fires, super storms, and flooding around the world, predicted years ago.

Fine said he watched a bear flee the wildfire, then attacked all but one of his goats, as he tells the story to show the collateral damage from the devastation.

“The destruction affects everything,” he continued. “Fires, floods, and water levels rising due to melting glaciers. All of this compels me to keep talking, keep teaching, and keep growing regenerative hemp. The good news is we have two new baby goats on the farm now, blessings abound!”

The ever-hopeful Fine explained that we don’t all have to become farmers, but we can begin to understand the process by growing a little patch of something—even if it’s a bunch of basil in a pot on a city balcony.

He does believe that farmers can lead the way, while being supported by the masses by small changes made to the way we live everyday.

“Supporting small, local farmers by buying locally-sourced products, getting produce from community-supported co-ops or farmer’s markets—or even working in community gardens, are all valuable contributions,” he surmised. “Who knows, you may find, like me, that farming or gardening and growing your own food is the most fun you’ll have outside the bedroom!”

These Bronze Age Humans Were History’s Biggest Weed Dealers

About 5,000 years ago, a mysterious tribe poured into Europe and Asia from the steppes of modern-day Russia and Ukraine. With them, they brought valuable knowledge of metalworking, horseback riding, and technology such as the wheel.

But historians believe these ancient migrants were also packing another precious commodity: none other than the finest marijuana the Caucasus had to offer. And according to new research, recently published in Vegetation History and Archaeobotany, it appears they were once the drug dealers of the entire Eurasian continent.

Anthropologists suspect the Yamnaya people were among the three or four prehistoric cultures that eventually founded Western civilization—flooding Europe with new languages, genetic admixture, and metal tools that marked the advent of the Bronze Age. At the very same time, they also ventured east, into China and central Asia.

See also  Bubble gum seeds

This tribe of nomadic herders originated in a territory called the Pontic-Caspian, and may have spoken Proto-Indo-European, the ancestor of hundreds of Indo-European languages used today. Some populations, such as Norwegians, can trace up to 50 percent of their DNA lineage to these ancient peoples.

Bronze age jewelry is renowned for its intricate gold working. Bracelets from a hoard in Woolaston, Gloucestershire. Image: Flickr

Even though the Yamnaya were prolific dope users, which scientists have discerned from large caches of archaeological evidence, weed has been cultivated throughout Eurasia for hundreds of thousands of years.

Mummified psychoactive marijuana, likely used in shamanistic rituals, has been discovered in the royal tombs of China’s Xinjiang region. Cannabis seeds dating back to 3,000 BCE were uncovered in the kurgan burial mounds of Siberia. And the tripped-out antics of the Scythians, a fearsome warrior clan, were once described by the Greek historian Herodotus as some “that no Grecian vapour-bath can surpass…transported by the vapor, [they] shout aloud.”

Yet, it wasn’t until the Bronze Age that people’s love of weed really started blazing. “The cannabis plant seems to have been distributed widely from as early as 10,000 years ago, or even earlier,” Tengwen Long, the study’s co-author, told New Scientist. But it would be another 5,000 years before signs of heavy marijuana use first appeared in archaeological records.

Map of the Hexi Corridor, also known as the Gansu Corridor. Image: Wikipedia

Despite the herb’s familiar presence in Eurasia, the study’s authors theorize that intercontinental trade during the Bronze Age helped crops like cannabis take hold in new places. Long before the Silk Road, another route known as the Hexi Corridor was favored by commodity traders, and possibly facilitated the flow of prehistoric ganja.

“It’s a hypothesis that requires more evidence to test,” Long admitted, but he also added that cannabis’ explosion throughout the continent suggests it might have been a “cash crop before cash.”

Since then, marijuana has been grown and domesticated by cultures all over the world. Today, various strains have been carefully selected for, and are responsible for the wide array of highs the drug is now capable of inducing.

But if it weren’t for the horse-riding, adze-wielding, herb-toking Yamnaya tribe, cannabis culture as we know it might not even exist. So go ahead, and light one up for these ancient ancestors who knew that a friend with weed is a friend indeed.


By signing up, you agree to the Terms of Use and Privacy Policy & to receive electronic communications from Vice Media Group, which may include marketing promotions, advertisements and sponsored content.

A classification of endangered high-THC cannabis (Cannabis sativa subsp. indica) domesticates and their wild relatives

This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Associated Data

Supplementary material 1: A classification of endangered high-THC cannabis (Cannabis sativa subsp. indica) domesticates and their wild relatives

Description of species concepts, level nominalism, wild-type nominalism, protologues, nomenclatural priority, intermediate forms, and elaboration of methodology. List of taxonomic characters and their respective coding, used in the morphological and total evidence analyses.

This dataset is made available under the Open Database License ( The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.


Two kinds of drug-type Cannabis gained layman’s terms in the 1980s. “Sativa” had origins in South Asia (India), with early historical dissemination to Southeast Asia, Africa, and the Americas. “Indica” had origins in Central Asia (Afghanistan, Pakistan, Turkestan). We have assigned unambiguous taxonomic names to these varieties, after examining morphological characters in 1100 herbarium specimens, and analyzing phytochemical and genetic data from the literature in a meta-analysis. “Sativa” and “Indica” are recognized as C. sativa subsp. indica var. indica and C. sativa subsp. indica var. afghanica, respectively. Their wild-growing relatives are C. sativa subsp. indica var. himalayensis (in South Asia), and C. sativa subsp. indica var. asperrima (in Central Asia). Natural selection initiated divergence, driven by climatic conditions in South and Central Asia. Subsequent domestication drove further phytochemical divergence. South and Central Asian domesticates can be distinguished by tetrahydrocannabinol and cannabidiol content (THC/CBD ratios, ≥7 or Cannabis that represent critically endangered reservoirs of germplasm from which modern cannabinoid strains originated, and which are in urgent need of conservation.


Cannabis is an ancient domesticate, a triple-use crop. Archaeologists found fruits (“seeds”) in a food context, a kitchen midden, with a calibrated radiocarbon date of 8000 cal BCE (Kudo et al. 2009). Evidence of fiber use is nearly as old, although identifying ancient cordage as Cannabis (or pottery impressions of same) is somewhat subjective (McPartland and Hegman 2018). Artifacts from a drug context-burnt residues with cannabinoids in a censer – date to 500 cal BCE (Ren et al. 2019). Early words for Cannabis include Chinese , attested ca. 750–600 BCE (Qu and Waley 1955), qunubu, a Neo-Assyrian loanword from the Scythian language, ca. 680 BCE (Seidel 1989), and κάνναβις, a Greek loanword from Scythian, ca. 440 BCE (Herotodus 2007).

The Latin name Cannabis sativa is usually attributed to Leonhart Fuchs, but the binomial was actually coined by Ermolao Barbaro, between 1480 and 1490, published 23 years after he died (Barbaro 1516). Carl Linné adopted the binomial in Species Plantarum, the internationally-recognized starting point of botanical nomenclature (Linnaeus 1753). Jean-Baptiste Lamarck broke from Linnaean orthodoxy by recognizing a second species, C. indica, for drug-type plants (Lamarck 1785).

Small and Cronquist (1976) proposed a single-species concept. They separated taxa by Linnaeus and Lamarck at the rank of subspecies, as C. sativa subsp. sativa and C. sativa subsp. indica (Lam.) E. Small & Cronq. The subspecies were circumscribed on the basis of ∆ 9 -tetrahydrocannabinol (THC) content. They defined C. sativa subsp. sativa as containing C. sativa subsp. indica as containing ≥0.3% THC. Numerous countries have incorporated the 0.3% criterion in regulations governing fiber-type (hemp) plants and drug-type (marijuana) plants.

Some botanists prefer to recognize C. sativa L. and C. indica Lam. at the rank of species (Hillig 2005a, Clarke and Merlin 2013). Debates over taxonomic rank are notoriously arbitrary. Molecular studies using DNA sequences can make the question of rank less arbitrary. Mandolino et al. (2002) quantified DNA polymorphisms in ten drug- and fiber-type varieties. They found more variability between individuals within a variety than between varieties – data that confirmed “the existence of a single, widely shared gene pool.” In a worldwide collection of Cannabis, Gilmore et al. (2007) found a low rate of sequence variation (approximately 1 polymorphism per 1 kb sequenced cpDNA) – consistent with a single species.

McPartland (2018) used DNA barcodes as a metric to place the Cannabis question of rank in context with other plants. He examined five plant barcodes (rbcL, matK, trnH-psbA, trnL-trnF, and ITS1), and calculated a mean divergence (barcode gap) of 0.41% between C. sativa and C. indica. This nearly equaled the mean divergence of 0.43% between five pairs of plants considered different varieties or subspecies (e.g., Camellia sinensis var. sinensis and C. sinensis var. assamica). In contrast, a 3.0% barcode gap separated five pairs of plants considered different species (e.g., Humulus lupulus and H. japonicus). Hebert et al. (2004) proposed a 2.7% difference between two COI sequences (the “barcode gap”) as the threshold for flagging genetically divergent specimens as distinct animal species.

Sawler et al. (2015) calculated a mean fixation index (FST) of 0.156 between populations of fiber- and drug-type plants (n = 43 and 81, respectively). FST values range from 0 to 1; a zero value indicates the two groups freely interbreed; a 1 value indicates the groups are completely isolated from one another. A mean FST of 0.156 is similar to the degree of genetic differentiation between human populations in Europe and East Asia, which belong to a single species.

Lynch et al. (2016) calculated FST = 0.099 between fiber- and drug-type groups (n = 22 and 173, respectively). Grassa et al. (2018) calculated FST = 0.229 between fiber-type accessions and “marijuana,” by concatenating data from Sawler, Lynch, and their own sequencing. Hey and Pinho (2012) proposed FST = 0.35 as a conservative threshold measure for species differentiation; pairs with greater values are identified as separate species, pairs with lesser values are identified as subspecies populations. Clearly, C. sativa L. and C. indica Lam. are best differentiated at a subspecies rank.

In the 1980s, drug-type plants came to be divided into two categories, widely known by the ubiquitous labels “Indica” and “Sativa”. This vernacular taxonomy became widespread after Anderson (1980) published a line drawing of the plants (Fig. ​ (Fig.1). 1 ). He differentiated “Indica” and “Sativa” by morphology and geographical provenance. As summarized by de Meijer and van Soest (1992), “Indica” applied to plants with broad leaflets, short and compact habit, and early maturation, and there is evidence that landrace ancestors of such plants came from Central Asia (primarily Afghanistan). “Sativa” applied to plants with narrow leaflets, tall and diffuse habit, and late maturation, and there is evidence that landrace ancestors of such plants came originally from South Asia (primarily India), with early historical distribution to Southeast Asia, Africa, and the Americas.

Line drawing adapted from Anderson (1980), courtesy of the Harvard University Herbaria and Botany Libraries.

Clarke (1981) accepted Anderson’s “Indica” concept for plants from Central Asia, “Strains from this area are often used as type examples for Cannabis indica.” In addition to morphological differences, he noted a phytochemical trait – Central Asian plants uniquely produced an acrid, skunk-like aroma. Clarke (1987) added an organoleptic quality – plants from Afghanistan produced a “slow flat dreary high.” Hillig (2005a) referred to Central Asian landraces as wide-leaflet diameter (WLD) biotypes, and landraces of South Asian heritage as narrow-leaflet diameter (NLD) biotypes. WLD and NLD biotypes differed in genetics (Hillig (2005a), morphology (Hillig 2005b), THC-to-cannabidiol (CBD) ratios (Hillig and Mahlberg 2004), and terpenoid content (Hillig 2004).

Recent authors have mistakenly equated the vernacular term “Sativa” with the epithet in the scientific name C. sativa, and mistakenly equated the vernacular term “Indica” with the epithet in the scientific name C. indica, mismatches first noted by McPartland et al. (2000). Small (2007) stated that “Sativa” and “Indica” were “quite inconsistent” with formal nomenclature. Linnaeus’s type specimen of C. sativa is a fiber-type (hemp) plant, not a drug-type (marijuana), and so the term “Sativa” has been inappropriately applied to drug-type plants (logically, it should be reserved for fiber-type hemp). Lamarck described C. indica for drug-type plants from India, and progenies in Southeast Asia and Africa – now counterintuitively called “Sativa” (logically, “Indica” should be reserved for the drug plants described by Lamarck).

See also  How to pick female marijuana seeds

The erroneous equivalences of vernacular “Sativa” (denoting plants with cannabinoids mostly or entirely THC) with “C. sativa” (in the narrow nomenclatural sense, denoting low-THC hemp forms), and vernacular “Indica” (denoting plants with substantial THC but also often substantial CBD) with “C. indica” (in the narrow nomenclatural sense, denoting high-THC, low-CBD forms) have appeared in taxonomic studies and legal documents. Even the pages of “Nature” have been problematically adorned with “Sativa” and “Indica”, accompanied by a version of Fig. ​ Fig.1 1 (Gould 2015). Those unfamiliar with the complexities and subtleties of biological classification can be misled, but in principle the issue is simple: the terms “Sativa” and “Indica” have been employed ambiguously and contradictorily.

In past centuries, landraces of South Asian heritage were grown over a much wider geographical range around the world than Central Asian landraces. The latter did not come to the attention of western Cannabis breeders until the early 1970s. Since then, breeders have haphazardly hybridized Central Asian and South Asian landraces, and largely obliterated their phenotypic differences (Clarke and Merlin 2013; Small 2017). Already 35 years ago, unhybridized landraces had become difficult to obtain in the USA and Europe (Clarke 1987). Hybrids of “Sativa” and “Indica” have proved overwhelmingly popular. “Indica” genes are useful for increasing cannabinoid yields, accelerating the maturity of outdoor plants at high latitudes, and reducing the height of plants so they are more easily concealed outdoors and more easily grown indoors. In the burgeoning CBD market, “Indica” genes (often from plants mislabeled “Ruderalis”) have increased the proportion of CBD relative to THC in plant products.

Alarmingly, Central and South Asian landraces have been corrupted by the introduction of foreign germplasm into their centers of diversity. Beisler (2006) boasted of importing “Mexican Gold” into Afghanistan around 1972. Casano (2005) noted that Afghani landraces were “disappearing” due to hybridization with other drug-type plants. Conversely, Central Asian landraces were introduced into South Asian centers of diversity in the 1970s – into Nepal (Cherniak 1982), Jamaica (Lamb 2010), and Thailand (Clarke and Merlin 2016). By 1980, Afghani landraces were imported into southern Kashmir, cultivated for sieved hashīsh, and escapes grew near crop fields (Clarke 1998). Also in the 1980s, Central Asian genetics were introduced into South Africa (Peterson 2009) and Morocco (Clarke and Merlin 2016). Sharma (1988) wrote about “hybrid Cannabis” growing in Kullu, Himachal Pradesh, and he implicated “foreign nationals.”

Central and South Asian landraces face extinction through introgressive hybridization. Wiegand (1935) first described this phenomenon in plants. Introgression refers to the infiltration of genes between taxa through the bridge of F1 hybrids. Fertile offspring from these crosses may display hybrid vigor (enhanced fitness), and replace one or both parental populations (Ellstrand 2003). Recent phylogenetic studies of populations allegedly representing “Indica” and “Sativa” show little or no genetic differences, because these studies primarily analyzed hybrid “strains” (Sawler et al. 2015; Dufresnes et al. 2017; Schwabe and McGlaughlin 2018). These results conflict with studies of landraces collected in the 1970s–1990s, which showed much clearer genetic differences (Hillig 2005a; Gilmore et al. 2007).

The use of “strain” names for IndicaSativa hybrids began with Watson (1985). A database of strain names currently lists 14,348 of them (Seedfinder 2019). This crowd-sourced enterprise – crossing and re-crossing hybrids of largely clandestine parentage – has resulted in a loss of genetic diversity (Mudge et al. 2018). Most strains sold by seed companies are characterized as “Sativa-dominant” or “Indica-dominant.” The arbitrariness of these designations is illustrated by “AK-47”, a hybrid strain that won “Best Sativa” in the 1999 Cannabis Cup, and won “Best Indica” four years later (McPartland 2017). Conceptually, a “strain” is equivalent to a “cultivar,” the latter being a taxonomic rank recognized by the “International Code of Nomenclature for Cultivated Plants” (ICNCP, Brickell et al. 2016). However, few commercial “strains” of drug-type Cannabis have met ICNCP requirements for cultivar recognition (Small 2015).

The ICNCP clusters cultivars into “Groups”. Consistent with ICNCP requirements, Small (2015) designated Central Asian landraces as “Cannabis Group Narcotic, THC/CBD Balanced,” and South Asian landraces as “Cannabis Group Narcotic, THC Predominant.” Some botanists argue that plants with traits created by human selection should be assigned cultivar status under the ICNCP, rather than assigned taxa under the “International Code of Nomenclature for Algae, Fungi, and Plants” (ICN, Turland 2018). However, for pragmatic reasons, botanists use the ICN framework to assign taxa to artificially selected plants (e.g., Hammer and Gladis 2014).

The above information has dealt basically with domesticated material. In addition, “wild” plants are also of concern. Cannabis “wild-type” traits were first described by Zinger (1898): small achene size, a persistent perianth with camouflagic mottling, and an elongated base – drawn out in the shape of a short, tapered stub with a well-developed abscission layer. In contrast, domesticated plants express a suite of phenotypic traits (the “domestication syndrome”) absent in wild-type plants, such as enlarged seed size, a lack of seed shattering (from reduction of the abscission zone), and reduction of perianth adherence.

Domesticated Cannabis easily escapes cultivation and goes “feral.” Domesticated C. sativa reverted to a wild-type phenotype in Canada just 50 generations (years) after cultivation was prohibited (Small 1975). This rapid phenotypic evolution makes it difficult to distinguish truly wild plants from formerly cultivated plants that have reverted to wild-type phenotypes. Thus Cannabis plants growing outside of cultivation could be (1) “volunteers” (escaped very recently from cultivation, maintaining their domesticated characteristics, and growing near where they were cultivated); (2) “escapes” that have readapted to wild existence (growing in various habitats, typically in disturbed or weedy places); or (3) “aboriginal” (unaltered by domestication and growing in their indigenous areas).

Aboriginal populations of several of the world’s most important crops do not seem to have survived, and Cannabis may be of this nature. Regardless, the wild-growing plants of Asia that are near (sympatric or parapatric) to the domesticates are of special significance. They may be direct ancestors of the domesticates, although this remains to be ascertained – many ancient domesticates were domesticated in locations distant from their sites of origin (Jarvis et al. 2016). In any event, there is considerable likelihood that the nearby wild plants of the domesticates share genes, since Cannabis produces massive quantities of pollen that is distributed for vast distances, and all Cannabis populations are capable of cross-pollination and completely interfertile (Small 1972). Accordingly, the wild varieties recognized in this publication represent very significant potential sources of genes representative of the endangered “Sativa” and “Indica” genomes.

This study does not address the European subspecies, C. sativa subsp. sativa. Small and Cronquist (1976) segregated this subspecies into two varieties – domesticated and wild-type plants. The domesticated variety is composed of fiber-type and oilseed landraces and cultivars. The wild-type variety has nomenclatural issues regarding C. sativa var. spontaneaVavilov (1922) and C. ruderalis (Janischevsky 1924). Vavilov and Janischevsky assigned these separate taxa to the same population of wild-type plants growing near Saratov, Russia. “Ruderalis” has become a mainstay of today’s vernacular taxonomy (Anderson 1980). See Suppl. material 1: SF.2 for a discussion of these nomenclatural issues, and an elaboration of “wild-type nominalism” in SF.3b.

Worldwide introgressive hybridization of “Indica” and “Sativa” threatens the agrobiodiversity of C. sativa. Seen pessimistically, the varieties described here are components of a vanishing world, and classifying them is like an exercise in renaming dinosaurs. Optimistically, the formal recognition of indigenous Central and South Asian varieties will provide them with unambiguous names, and may help prevent their extinction.


Taxonomic characters for analysis included aspects of morphology, phytochemistry, genetics, and host-parasite relationships. Some data are new (morphological studies of herbarium specimens), whereas phytochemical and molecular data were extracted from previously published studies. Most of those studies employed common garden experiments (CGEs). CGEs grow plants from different places in a single location, under common environmental conditions, with uniform processing (Grassi and McPartland 2017).

Morphological characters

Approximately 1,100 herbarium specimens were examined, at 15 herbaria, designated by herbarium acronyms in Index Herbariorum (Suppl. material 1: SF.4). Additionally, we extracted morphological data from CGEs that compared Central and South Asian germplasm collected in the previous century (e.g., Vavilov and Bukinich 1929, Small et al. 1976, Anderson 1980, de Meijer 1994, Hillig 2005b). We also drew on morphological data from archaeobotanical studies. In the spirit of open access, extracted morphological data are provided in Suppl. material 1: SF.8, permitting readers to synthesize the raw data for themselves. CGE studies provided data often absent in herbarium specimens, such as plant height, internode length, stalk thickness, and branch angle or divarication.

Branch angle or divarication measured the angle, in degrees, that a branch came off the vertical shoot; it generally ranged between 35° to 85° from vertical. Branch angle may be a function of internode length, which was also assessed. Branch flexibility is a qualitative measure of the ability of a branch to bend or droop without snapping. Flexibility likely reflects the ratio of bast fiber (flexible) to wood fiber (inflexible). Leaf morphology was assessed in “fan leaves” (i.e. larger palmately compound leaves) near the base of inflorescences. The sampled leaves conformed to the concept of 1 st order branching off the main shoot, as presented by Spitzer-Rimon et al. (2019). Central leaflet length/width ratio (L/W) is expressed as a quotient. Leaflet shape was either lanceolate (the widest part is less than midway down the length of the leaflet from its base), or oblanceolate (where the widest location is more than half way down the length). This was measured as the distance to the widest point (WP) divided by the entire length (WP/L). A leaflet with WP/L > 0.5 is oblanceolate (Anderson 1980).

The perigonal bract (also called bracteole, perigonium, or inappropriately “calyx”) is the floral bract enclosing the female flower and later the achene (Small 2015). Inflorescence density was qualitatively assessed using the “perigonal bract-to-leaf index” (i.e., the “calyx-to-leaf ratio,” Clarke 1981). Inflorescences with a low index have a predominance of leaf material – interstitial “sugar leaves” (relatively small leaves with few leaflets occurring in the inflorescence) between clusters, subtending 2 nd order to 7 th order branchlets (Spitzer-Rimon et al. 2019). A low index is associated, in part, with short internode length and broad leaflet width.

See also  Identification of marijuana seeds

The density of capitate-stalked glandular trichomes (CSGTs) was qualitatively assessed (i.e. visually evaluated) on perigonal bracts. CSGT density was mentioned by Christison (1850) in one of the first CGEs that compared C. sativa (Scottish hemp) and C. indica (Indian gunjuh). He noted that C. indica inflorescences felt resinous when touched, “Floral leaves, bracts, and perianth covered with glandular pubescence.” He also noted that C. indica leaves produced “both sessile glands and glandular hairs [CSGTs].” CSGT density on sugar leaves was also qualitatively assessed, based on the method by Potter (2009).

As used here, the “fruit” includes the achene and its more or less adherent perianth. In female flowers of Cannabis, the perianth does not produce a corolla, but instead adheres to the exocarp (outermost layer of the achene wall). Dimensions and appearance of the fruit were assessed.

For each herbarium specimen, a standardized form was used to record specimen label data (collector name, date, location, annotations) and morphological data. During the course of this study, morphological characters were added (e.g., branch angle, inflorescence density, CSGT density), necessitating return visits to some herbaria (BM, ECON, GH, IND, K). Morphological data were synthesized qualitatively (e.g., branch flexibility, leaf color, inflorescence density, CSGT density, perianth adherence), or quantitatively (e.g., plant height, internode length, leaflet L/W and WP/L ratios, achene size). Quantitative data provided bracket measurements for each described taxon.

Phytochemical characters

A widely-cited paper by Turner et al. (1980) listed 420 phytochemicals isolated from C. sativa – the 420 plant. Few phytochemicals provide useful taxonomic information, however. Our study focused on cannabinoids and terpenoids. In living plants and freshly harvested tissues, cannabinoids exist predominantly in the form of carboxylic acids. THC occurs as tetrahydrocannabinolic acid (THCA); cannabidiol (CBD) occurs as cannabidiolic acid (CBDA). Decarboxylation of the cannabinoids into their neutral counterparts occurs relatively slowly with aging, and rapidly with heat. Thus THCA converts to THC, and CBDA converts to CBD. In addition, when THC ages (unless appropriately stored) it substantially transforms to cannabinol (CBN), an oxidation product. In this paper when THC and CBD are mentioned it should be understood that depending on context, “THC” may mean THCA + THC + CBN, and “CBD” may mean CBDA + CBD.

Rather than cannabinoid quantity (i.e., THC% w/w), we report a parameter measuring cannabinoid quality: the THC/CBD ratio (THC% w/w divided by CBD% w/w). The THC/CBD ratio is a quite conservative (stable) character, whereas THC% correlates with morphology, such as trichome density (Potter 2009), as well as inflorescence density and gland head size. These morphological differences do not alter the THC/CBD ratio. The ratio is determined by a single gene with codominant alleles (de Meijer et al. 2003), or two tightly-linked yet separate THCAS and CBDAS genes (Van Bakel et al. 2011, Laverty et al. 2019). Weiblen et al. (2015) identified a single quantitative trait locus (QTL) associated with the THC/CBD ratio.

In contrast, THC% expression is polygenic, altered by many genes that contribute to morphological differences. Environmental factors (light intensity, temperature, soil nutrients, etc.) alter THC%, but have much less effect on THC/CBD. As a dimensionless ratio, THC/CBD provides a more valid comparison of many studies that grew plants under different conditions (Grassi and McPartland 2017).

Tetrahydrocannabivarin (THCV) and cannabidivarin (CBDV) are short-tailed C19 analogs of THC and CBD. The biosynthetic pathway leading to THCV and CBDV diverges early, on the resorcinol side of the cannabinoid pipeline. Some researchers add C19 analogs to THC/CBD ratios, as THC+THCV/CBD+CBDV (e.g., Turner et al. 1980). Here, the percentage of C19 analogs (THCV%+CBDV%) is treated as a separate character.

Terpenoids constitute the “essential oil” of Cannabis. Terpenoids include hydrocarbon terpenes and their oxygenated derivatives, which form alcohols, ethers, aldehydes, ketones, and esters. They are volatile, and give the plant its characteristic smell. Christison (1850) noted that Indian gunjuh emitted a balsamic odor, lacking in Scottish hemp. South Asian landraces often smell “herbal” or “sweet,” whereas Central Asian landraces give off an acrid or “skunky” aroma (Clarke 1981).

Genetic characters

Molecular genetic studies of Central and South Asian populations – which have not been significantly hybridized in recent times – are limited in number. Twenty years ago, when unhybridized landraces were much more readily available, molecular methods were blunt instruments. Today, we can decode the DNA sequence of whole genomes, but a good representation of the range of unhybridized biodiversity is not available for analysis, although collection of genuinely representative germplasm from Asia may still be possible. Herbaria of course are invaluable repositories of older specimens, but collections from Asia are relatively limited, and for various reasons, curators have often been unable to allow sampling of older collections.

Herbarium voucher specimens were deposited for some CGE studies (Small and Beckstead 1973; Turner et al. 1973, 1979; de Meijer et al. 1992; de Meijer 1994; Hillig 2004, 2005a; Hillig and Mahlberg 2004; Gilmore et al. 2007), which we examined to ascertain correlations with morphology. For other phytochemical and genetic studies, we relied upon reports of geographic provenance of their accessions.


The electronic version of this article in Portable Document Format (PDF), in a work with an ISSN or ISBN number, represents a published work according to the ICN (Turland 2018). Hence the new names contained in the electronic publication of this article are effectively published under the ICN from the electronic edition alone. New names contained in this work have been submitted to the International Plant Names Index (IPNI,, from where they will be made available to the Global Names Index.

An example of a taxonomic trait shifting over the past 50 years, as Central Asian landraces hybridized into “Indica”, is provided in Fig. ​ Fig.2. 2 . It illustrates a convergence in THC/CBD ratios over the past 50 years. In studies of accessions collected in the 1970s–1990s, Central Asian landraces (study numbers in unitalicized red font), the THC/CBD ratio, expressed as a quotient, was always < 7 (study size weighted mean = 3.56). In studies of South Asian landraces collected in the 1970s–1990s (study numbers in italicized green font), the THC/CBD ratio was ≥ 7 (study size weighted mean = 97.14). Since then, THC/CBD ratios have skyrocketed in accessions purportedly representing Central Asia (i.e., “Indica”). Now there is little or no difference between “Indica” and “Sativa”.

Shifts in THC/CBD ratios over time; data from 47 numbered studies in Suppl. material 1: SF.9. Central Asian landraces in unitalicized red (n =13 studies); “Indica” in underlined unitalicized red (n= 9); South Asian landraces in italicized green (n =18 studies); “Sativa” in underlined italicized green (n =7 studies). Size of numeral reflects the number of accessions analyzed in that study.

Taxonomic analysis

We classified C. sativa subsp. indica into four varieties (in the formal nomenclatural sense, i.e., varietas). Two varieties express traits of domestication (identical to “Indica” and “Sativa” in the original narrow meanings of these terms), and two varieties have wild-type traits. We followed precedent set by Small and Cronquist (1976) who segregated C. sativa subsp. indica into two varieties – domesticated and wild-type plants. They did not place these varieties in an ancestor–progeny relationship, however, because they could not verify putative ancestral relationships.

Key to four varieties of C. sativa subsp. indica 1

1. Plants usually with a THC/CBD ratio ≥7; terpenoid profile usually lacks sesquiterpene alcohols, fresh aroma often pleasant. Plants ≥ 2 m tall in good habitats; branches flexible, diverging from the shoot at a relatively acute angle (6), lanceolate to linear-lanceolate; margins with fine to coarse serrations, sometimes biserrate. Mature female inflorescence somewhat compact (flowering stems producing small to medium “buds”), with relatively obscure sugar leaves (a high perigonal bract-to-leaf index); sugar leaves with capitate-stalked glandular trichomes (CSGTs) usually limited to the proximal half of the leaves; perigonal bracts express a moderate to high density of CSGTs. Mature achene exocarp color (beneath the perianth) often green-brown.
A THC/CBD ratio always ≥7, often much more. Mature achenes usually ≥ 3.6 mm long (Fig. 3e, f ); perianth mostly sloughed off, but often persistent in places (appearing as irregular spots or stripes); exposed exocarp exhibiting prominent venation; lacking a prominent protuberant base; not readily disarticulating from plant var. indica (“Sativa” in the historical sense 2 )
B THC/CBD ratio usually ≥7, sometimes less. Mature achenes usually

var. himalayensis
2. Plants with a THC/CBD ratio ca. 1 m; branches not flexible, branching sometimes nearly 90° from the stalk axis, producing a menorah-shaped habitus. Fresh leaves dark green in color, leaflets of larger leaves sometimes overlap; central leaflets broad (length/width usually <6), often oblanceolate; margins with coarse serrations, rarely biserrate. Mature female inflorescence compact (flowering stems producing medium to large “buds”) with prominent sugar leaves (a low perigonal bract-to-leaf index); sugar leaves have CSGTs extending more than half way down their length; perigonal bracts densely covered with CSGTs. Mature achene exocarp color (beneath the perianth) often a lighter shade of olive green to gray.
A THC/CBD ratio 2). Mature achenes usually ≥ 3.6 mm long (Fig. 3a, b ); perianth mostly sloughed off (appearing as irregular spots or stripes); exposed exocarp exhibiting prominent venation; lacking a prominent protuberant base; not disarticulating from plant, and often trapped in the dense inflorescence var. afghanica (“Indica” in the historical sense 2 )
B THC/CBD ratio often

var. asperrima
1 As emphasized in the text, the differences presented here represent unhybridized plants, before extensive recent hybridization between them.
2 Historically, as discussed in the text, “Sativa” formerly represented landraces of South Asian heritage, and “Indica” formerly represented Central Asian landraces. This key is not intended for the identification of “Sativa” and “Indica” strains commercially available today.

Representative achenes of four varieties Aindica, Rajshahi (Bangladesh), Clarke 1877 (BM) Bindica, Coimbatore (India), Bircher 1893 (K) Cindica, South Africa, Hillig 1996; (IND) Dhimalayensis neotype Ehimalayensis, Bareilly (India), Roxburgh 1796 (K). Fhimalayensis, East Bengal (Bangladesh) Griffith 1835 (GH) Gafghanica neotype Hafghanica epitype Iafghanica Yarkant (Xīnjiāng), Henderson 1871 (LE) Jasperrima lectotype Kasperrima Nuristān (Afghanistan), Street 1965 (F) L Kailiyskiy Alatau (Kazakhstan), Semenov-Tyan-Shansky 1857 (LE).

Taxonomic treatment

Please note that light quality varied among herbaria, so photographs of herbarium specimens and achenes at different herbaria varied somewhat in their tint, hue, and tone. For protologues of the four varieties (everything associated with a basionym at its time of publication), see Suppl. material 1: SF.6. For additional representative herbarium specimens of the four varieties, see Suppl. material 1.