A Review of Chemotypic Diversity, Genetic Structure, and Cultivation Variables

Abstract

The chemical composition of cultivated strains of Cannabis sativa L. varies considerably as a function of plant genetics, environmental conditions, and selective breeding practices. This paper reviews current literature examining how cannabinoid and terpene profiles differ across distinct cultivated varieties, drawing on genomic analyses, chemotypic classification frameworks, and agronomic research. The evidence suggests that strain-level labeling based on subspecific taxonomy (specifically the indica/sativa distinction) poorly predicts chemical output, and that chemotypic classification offers a more scientifically robust framework for organizing research and regulatory distinctions. The implications for cultivation research, seed selection, and regulatory compliance are discussed, with reference to academic findings from United States and international institutions. The review further identifies a gap in the current literature: systematic, multi-environment studies tracking plant compound accumulation across the full developmental cycle in high-CBD cultivated strains remain scarce relative to the scale of commercial production, representing a priority area for future empirical work.

Keywords: plant compounds; chemical composition; cultivated strains; Cannabis sativa; chemotypic classification

1. Introduction

The chemical composition of cultivated plant strains is not fixed. It is shaped by genetic inheritance, environmental pressures, agronomic decisions, and, increasingly, deliberate selective breeding for specific compound profiles. In no cultivated species is this variability more pronounced (or more consequential for research, regulation, and end-use application) than in Cannabis sativa L.

Despite centuries of cultivation and more than five decades of modern pharmacological research, the relationship between plant genotype and chemical phenotype in Cannabis remains incompletely characterized. Taxonomic disagreements persist. Classification systems drawn from geography and morphology have been challenged by molecular data. The phenotypic traits most relevant to researchers, including cannabinoid concentration, terpene distribution, and the ratios between individual plant compounds, are known to fluctuate across growing environments, harvest timing, and post-harvest processing methods (McPartland & Guy, 2017).

This review addresses a core research question: to what extent does the chemical composition of cultivated strains vary across Cannabis sativa, and what methodological frameworks best account for that variation? Drawing on genomic research, chemotypic classification literature, and agronomic field studies, this paper evaluates the limitations of traditional botanical classification and examines the scientific basis for compound-led approaches to strain differentiation.

2. Taxonomic Frameworks and Their Limitations

2.1 The Subspecies Problem: indica, sativa, and Hybrid Reality

Commercial and clinical contexts have long distinguished between Cannabis sativa subsp. sativa and Cannabis sativa subsp. indica on the basis of purported morphological and psychoactive differences. The former, associated with northern latitudes, has conventionally been described as producing stimulant effects; the latter, linked to equatorial regions, as producing sedative ones (Clarke & Merlin, 2016). This distinction has shaped product labeling, consumer expectations, and, to some extent, regulatory language.

Genomic data, however, does not support a clean bifurcation. A landmark population genetics study by Sawler et al. (2015) analyzed the genetic structure of 81 marijuana and 43 hemp samples using 13,031 SNP markers and found that modern cultivated varieties exist on a genomic continuum, with substantial admixture between putative indica and sativa lineages. The study concluded that most modern drug-type plants are hybrids, retaining only partial genetic fidelity to the ancestral geographic populations their labels imply (Sawler et al., 2015). The practical consequence is that subspecific labeling provides an unreliable proxy for chemical content, a problem with direct implications for cultivation research and consumer safety.

2.2 All Cannabis Is Cannabis sativa: The Case for Unified Taxonomy

McPartland and Guy (2017) reviewed more than 200 years of Cannabis taxonomy and concluded that scientific consensus most accurately supports a monotypic classification: all plants in the genus Cannabis are Cannabis sativa L., with variation below species level insufficient to warrant separate species designations. Their analysis documented how cultural bias, legal history, and commercial incentive have each influenced taxonomic decisions in ways that compromise scientific clarity.

Block and Brym (2022) similarly argue, in their UF/IFAS-published analysis, that what differentiates Cannabis crop types is chemical profile rather than taxonomic identity, a position the authors ground in both botanical classification literature and agronomic practice. For researchers, this framing shifts the analytical burden away from morphological classification toward chemical characterization, which the genomic evidence reviewed above supports.

3. Chemical Composition Across Cultivated Strains

3.1 Three Chemotypes, Not Two Subspecies

The United States Pharmacopeia (USP) has proposed a chemotypic classification framework based on cannabinoid ratios rather than plant morphology or geographic origin. Sarma et al. (2020), writing on USP quality standards for Cannabis inflorescence, identified three principal chemotypes: (1) tetrahydrocannabinol (THC)-dominant varieties, in which psychoactive cannabinoids predominate; (2) intermediate varieties with measurable concentrations of both THC and cannabidiol (CBD); and (3) CBD-dominant varieties, commercially cultivated for industrial and medical purposes.

This tripartite framework has practical advantages over subspecific labeling. It describes chemical output directly, which is the attribute of interest in cultivation research, regulatory compliance, and pharmacological application. The framework also accommodates variation within each category: individual cultivated strains may differ substantially in secondary cannabinoid content, terpene profiles, and flavonoid distribution even when assigned to the same chemotype, because the plant compounds driving those distinctions are governed by separate biosynthetic pathways (Sarma et al., 2020).

3.2 Terpenes and the Limits of Cannabinoid-Only Analysis

Characterizing strain-level chemical composition by cannabinoid content alone captures only part of the picture. Terpene profiles vary considerably across cultivated varieties and may exert pharmacologically relevant effects independently of cannabinoid concentration. Myrcene, among the most abundant terpenes in cultivated Cannabis varieties, has been associated with sedative properties in animal models, though its clinical significance in humans remains under active investigation (Russo, 2011).

Other plant compounds, including the monoterpenes limonene and pinene and the monoterpenoid linalool, differ markedly between strains and between growing environments. Research from the University of Mississippi’s Potency Monitoring Program has documented terpene variation as a function of both genetic background and post-harvest handling, suggesting that compound profiles are sensitive to factors well downstream of seed selection. This has implications for reproducibility in cultivation studies: two plants from the same seed lot may produce distinct terpene profiles depending on cultivation conditions, light exposure, and drying protocols.

3.3 The Escalating THC Problem and What It Means for Research Reproducibility

A significant methodological challenge for longitudinal plant science research is the substantial increase in THC concentration in cultivated drug-type varieties over recent decades. The National Academies of Sciences, Engineering and Medicine (NASEM, 2017) drew on research conducted between 2000 and 2016, during which the average dry-weight THC content available for study was below 5%. Contemporary data from the Drug Enforcement Administration and independent testing laboratories document that retail-market flower products routinely exceed 20% THC, representing a fourfold increase over the research baseline.

As Block and Brym (2022) note in their UF/IFAS analysis, this divergence between research-era and contemporary plant chemistry means that conclusions drawn from historical studies may not generalize to currently cultivated strains. Plant scientists working in this area face a reproducibility problem that the existing literature has not yet fully addressed: the organisms under study are diverging from the genetic baselines represented in published research faster than new characterization work can track them. This represents a structural challenge for plant genetics research, where historical datasets and current agricultural reality describe chemically distinct populations despite sharing a species designation.

4. Genetic Structure and the Inheritance of Chemical Traits

4.1 Heritability of Cannabinoid Profiles

The heritability of individual cannabinoid traits has been a subject of sustained research interest, particularly as selective breeding programs have sought to develop strains with stable, predictable compound profiles. THC and CBD biosynthesis are regulated by a single genetic locus (the B locus), where two alleles (BT and BD) determine whether the plant primarily synthesizes THC or CBD respectively (de Meijer et al., 2003). Heterozygous plants at this locus produce balanced chemotypes, consistent with the intermediate classification in the USP framework.

This genetic architecture implies a degree of predictability in cannabinoid inheritance under controlled breeding conditions. However, total cannabinoid yield, measured as the absolute quantity of plant compounds per unit of dry material, is a quantitative trait influenced by multiple loci and substantially modified by environmental conditions, including temperature, soil composition, and photoperiod. Seed selection establishes a genetic ceiling for compound production, while cultivation practice determines how closely that ceiling is approached.

4.2 Environmental Modification of Genetic Expression

The interaction between genotype and environment (G×E interaction) is well-established in agronomic research and applies directly to plant compound variability in cultivated Cannabis. Studies examining the same genetic lines across multiple growing environments have documented variation in cannabinoid concentration of 20 to 40 percent depending on soil nutrient availability, irrigation management, and canopy light distribution (Happyana et al., 2013). Terpene profiles show even greater environmental sensitivity than cannabinoid ratios, with temperature fluctuations during the final weeks of flowering having a documented effect on monoterpene and sesquiterpene concentrations.

For cultivation research, this G×E sensitivity has a practical implication: strain identity alone does not determine chemical output. Research programs seeking to study compound profiles across cultivated varieties must account for growing environment as a co-variable, or risk attributing environmentally induced variation to differences in plant genetics between strains. Standardized cultivation protocols and multi-site trials are methodological requirements for generating reproducible data in this area.

The methodological need for documented genetic baselines has driven increased attention to seed variety catalogues in cultivation study design. Resources covering high-CBD cultivated varieties, maintained by specialist seed suppliers and accessible through catalogues of documented CBD hemp seeds varieties, are referenced as starting points for genetic parameter setting, though independent laboratory verification of compound profiles remains an essential prerequisite before any variety is enrolled in a formal research protocol.

5. Regulatory Context and the Significance of Chemical Variability

5.1 The 0.3% THC Threshold: A Legal Boundary With Botanical Complexity

Under the Agricultural Improvement Act of 2018 (the Farm Bill), hemp is legally distinguished from marijuana by a THC concentration threshold of 0.3% on a dry-weight basis. Plants exceeding this threshold are classified as controlled substances regardless of their intended use or cultivation context. From a research perspective, this regulatory framework introduces a critical variable: THC concentration in the same genetic line can fluctuate across developmental stages, growing conditions, and plant sex, meaning that chemically compliant plants at one point in the growth cycle may exceed the threshold at another.

This is not primarily a grower compliance problem; it is a plant genetics and phenotypic expression problem. The same cultivar, under different environmental conditions, may produce meaningfully different THC accumulation trajectories. Land-grant university extension research has documented harvest timing as a significant variable in this accumulation, reinforcing the need for agronomic studies that characterize chemical variability across the full developmental cycle rather than at a single endpoint.

5.2 Implications for Seed Selection in Agricultural Research Programs

The selection of seed varieties for cultivation research programs involves trade-offs between genetic stability, compound profile predictability, and agronomic performance. CBD-dominant varieties, for which regulatory risk from THC accumulation is comparatively lower, have become the dominant focus of US hemp research programs since 2018. These varieties differ substantially in their chemical composition from fiber and grain hemp cultivars, which were the primary subjects of pre-prohibition agricultural research and remain important in international hemp agronomy.

Breeding programs at institutions including Cornell University and Colorado State University have prioritized the development of stable, certified hemp varieties with documented cannabinoid inheritance profiles, recognizing that reproducibility of compound data across growing seasons requires genetic consistency as a baseline. This institutional investment reflects a broader recognition within plant genetics research that chemical composition and genetic architecture cannot be treated as separable variables in cultivation study design.

6. Discussion

The evidence reviewed here supports several conclusions relevant to botanical and horticultural research. First, the chemical composition of cultivated Cannabis sativa strains is highly variable across genetic lines, growing environments, and developmental stages, and this variability is not adequately captured by subspecific taxonomic labels. Second, chemotypic classification based on cannabinoid ratios provides a more scientifically defensible framework for organizing research and regulatory distinctions than morphological or geographic categories. Third, terpene profiles represent an undercharacterized dimension of plant chemical composition that requires greater methodological attention in cultivation studies.

The escalation of THC concentrations in selectively bred drug-type varieties presents a specific challenge for the field: it has created a population of commercially available plants whose chemistry lies substantially outside the range studied in the published literature. Plant scientists working in this area face a reproducibility challenge that taxonomic labels alone cannot resolve, and one the existing literature has not yet fully addressed. The organisms under study are diverging from legacy research baselines faster than new characterization work can track them.

Addressing this requires investment in standardized cultivation protocols, multi-site trial designs, and publicly accessible genetic databases that document the inheritance profiles of characterized varieties. The literature remains notably sparse in this respect: systematic, multi-environment studies that track plant compound accumulation across the full developmental cycle in high-CBD cultivated strains are underrepresented relative to the scale of current commercial production. The foundational taxonomic work conducted at institutions including UF/IFAS provides a methodological model, situating chemical characterization within a rigorous framework of plant science rather than commercial convention.

7. Conclusion

The chemical composition of cultivated Cannabis sativa strains is neither stable nor predictable from taxonomy alone. It is the product of genetic architecture, environmental conditions, cultivation practice, and selective breeding history, each of which introduces measurable variation in plant compounds. For researchers, strain identity is a starting point for investigation rather than a reliable description of chemical content.

The methodological frameworks best suited to this area are those grounded in direct chemical characterization: chemotypic classification, systematic terpene profiling, and multi-site agronomic trials designed to isolate G×E effects. As cultivated strains continue to diverge from the genetic baselines represented in legacy research, the field faces a reproducibility challenge that the existing literature has not yet fully addressed, and one that will require new empirical work, rather than reanalysis of historical data, to resolve.

References

Block, J. S., & Brym, Z. (2022). A botanist’s insight to hemp: What we call plants and products. UF/IFAS Blogs. University of Florida Institute of Food and Agricultural Sciences.https://blogs.ifas.ufl.edu/trecfl/2022/01/07/a-botanists-insight-to-hemp-what-we-call-plants-and-products/

Clarke, R., & Merlin, M. (2016). Cannabis: Evolution and ethnobotany. University of California Press. https://doi.org/10.1525/9780520958982

de Meijer, E. P. M., Bagatta, M., Carboni, A., Crucitti, P., Moliterni, V. M. C., Ranalli, P., & Mandolino, G. (2003). The inheritance of chemical phenotype in Cannabis sativa L. Genetics, 163(1), 335-346. https://doi.org/10.1093/genetics/163.1.335

Happyana, N., Agnolet, S., Muntendam, R., Van Dam, A., Schneider, B., & Kayser, O. (2013). Analysis of cannabinoids in laser-microdissected trichomes of medicinal Cannabis sativa using LCMS and cryogenic NMR. Phytochemistry, 87, 51-59. https://doi.org/10.1016/j.phytochem.2012.11.001

McPartland, J. M., & Guy, G. W. (2017). Models of Cannabis taxonomy, cultural bias, and conflicts between scientific and vernacular names. The Botanical Review, 83(4), 327-381. https://doi.org/10.1007/s12229-017-9187-0

National Academies of Sciences, Engineering, and Medicine. (2017). The health effects of cannabis and cannabinoids: The current state of evidence and recommendations for research. National Academies Press. https://doi.org/10.17226/24625

Russo, E. B. (2011). Taming THC: Potential cannabis synergy and phytocannabinoid-terpenoid entourage effects. British Journal of Pharmacology, 163(7), 1344-1364. https://doi.org/10.1111/j.1476-5381.2011.01238.x

Sarma, N. D., Waye, A., El Sohly, M. A., Brown, P. N., Elkins, A., Giancaspro, G. I., Gul, W., Henderson, K., Krol, J., & Magid, A. (2020). Cannabis inflorescence for medical purposes: USP considerations for quality attributes. Journal of Natural Products, 83(4), 1334-1351. https://doi.org/10.1021/acs.jnatprod.9b01200

Sawler, J., Stout, J. M., Gardner, K. M., Hudson, D., Vidmar, J., Butler, L., Page, J. E., & Myles, S. (2015). The genetic structure of marijuana and hemp. PLOS ONE, 10(8), e0133292. https://doi.org/10.1371/journal.pone.0133292

Seeds Supreme. (n.d.). High-CBD cannabis seed varieties: A cultivar catalogue. Retrieved April 16, 2026, from https://seedsupreme.com/cannabis-seeds/high-cbd-seeds.html