Dahlia Doctor Research Library: Dahlia Cultivar Evaluation, Classification, and Trait Standards

A Dahlia Doctor Research Library Collection 

Copyright © 2026 by Steve K. Lloyd
All Rights Reserved

A dahlia cultivar is more than a name and a color. It is a set of traits: leaf arrangement, stem caliper, flower diameter, ray-floret geometry, color chart readings, developmental rate, and dozens of other measurable characteristics that define how a plant grows, blooms, and differs from its neighbors in a trial plot.

Making those traits measurable, comparable, and repeatable across sites and seasons is the work this collection documents.

The research gathered here spans morphological descriptor systems developed for plant-variety protection, statistical methods for distinguishing real genetic differences from environmental noise, multivariate approaches to cultivar grouping, and phenological models that help explain why the same cultivar can perform differently across locations and seasons. The aim is not to rank cultivars or recommend varieties. It is to show how researchers turn the observable features of a dahlia into data that can be compared, inherited, and interpreted.

One boundary is worth noting at the outset. Horticultural classification systems organize cultivars by visible flower form, but molecular evidence shows that those form-based groupings do not necessarily reflect genome-wide genetic relationships. That distinction appears in the final cluster of this collection, where it belongs: as a caution about what classification systems can and cannot tell us about genetic identity.

Companion collections in the Dahlia Doctor Research Library cover flower color genetics, breeding systems, mutation and sports, wild species diversity, cut flower production, and vase life. Those collections address genetic mechanisms, hybridization, production performance, and postharvest behavior in greater depth. This collection focuses on the methods researchers use to characterize and compare cultivars as phenotypes, meaning as sets of measurable and observable traits.

Each post in this series presents a curated set of Dahlia Doctor Knowledge Cards organized around a specific research topic. A Knowledge Card summarizes one scientific or technical source using a consistent structure: study system, experimental context, experimental design, key results, mechanistic insight, practical guidance, and why the source matters to dahlia growers and researchers.

These summaries represent original interpretive work. They are intended as a research guide, not a substitute for reading the original papers. Each citation title links to a Google Scholar search or direct source link, opening in a new tab when possible, to help you locate the original publication independently.

Each Knowledge Card appears once in this collection, placed in the topic cluster where it contributes most directly. Some sources are relevant to more than one cluster. Placement reflects primary emphasis rather than exclusive relevance.

This collection covers cultivar evaluation, morphological descriptors, trait scoring, distinctness-uniformity-stability characterization, phenotypic diversity, character association, multivariate grouping, and the limits of classification systems. It does not recommend specific cultivars. When a source reports that one named cultivar, accession, or genotype performed best in a particular trial, that result is treated as site-specific evidence about trait expression under defined conditions, not as a general buying or growing recommendation.

KC-0932 (Thakur et al., 2026) was previously used in the Dahlia Doctor Research Library collection on Dahlia Structure and Anatomy, where it supported morphological and anatomical description. Here it serves as the formal descriptor-standard anchor for cultivar characterization, with emphasis on its distinctness-uniformity-stability framework, fifty-one-trait protocol, and hierarchical cluster output. The Why This Source Matters framing in this entry reflects its role in the current collection.

KC-0792 (Wegner & Debener, 2008) was previously used in the Dahlia Doctor Research Library collection on Dahlia Breeding Systems and Polyploid Genetics. Its role here is narrower. The AFLP molecular analysis by Wegner and Debener showed that genetic clustering among dahlia cultivars and wild species did not correspond to horticultural classification categories based on flower form. This source is used here only as evidence for the limits of morphology-based classification, not as a general breeding-strategy recommendation.

KC-0094 (Fernandes et al., 2023) and KC-0237 (Fernandes et al., 2024) are companion papers from the same research group at Universidade Federal de Santa Maria, studying phyllochron and developmental dynamics in on-farm cut dahlia. They have not appeared in prior Dahlia Doctor Research Library collections. They are included here not as production-timing recommendations, but as evidence that cultivar trait expression is environmentally conditioned. Their role is to support the argument that trial results from one site or season do not transfer cleanly to other growing conditions.

The source catalogued as KC-0247 (Li, 2019) carries the title “Landscape Appreciation of Fifty Dahlia Species in Xi’an Area,” but the study evaluates cultivars of Dahlia pinnata rather than wild species. The word “species” in the title appears to reflect translation convention rather than taxonomic usage. The study system is cultivated dahlia.

The starting point for rigorous cultivar evaluation is a shared vocabulary of observable traits. Descriptor systems provide that vocabulary. They specify which plant and floral characteristics to measure, how to measure them, and how to record the result in ways that allow one evaluator’s observation in one location to be compared with another’s observation elsewhere.

The three sources in this cluster document what those descriptor systems look like in practice for dahlias: how populations of cultivars differ across defined traits, what the variance structure of that difference looks like, and how correlations among morpho-decorative traits operate within a defined flower-form class.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Thakur, P., Sadhukhan, R., Choudhary, T., Srivastava, M. K., & Kashyap, S. (2026). DUS characterization of dahlia (Dahlia variabilis). Indian Journal of Plant Genetic Resources, 39(1), 100–104.

Study System

Thirty-two Dahlia variabilis genotypes grown at a regional horticultural research station in a low-hill subtropical setting in India.

Experimental Context

Morphological characterization of dahlia germplasm for distinctiveness, uniformity, and stability profiling. Vegetative, floral, flowering-duration, vase-life, and tuber-shape traits were recorded for varietal identification and grouping.

Experimental Design

Terminal cuttings from healthy tubers were rooted after treatment with indole-3-butyric acid at 500 to 1000 parts per million. Rooted cuttings were transplanted to field plots at 60 cm by 40 cm spacing. The trial used three replications, with observations recorded on ten randomly selected plants per replication over two consecutive growing seasons. Fifty-one morphological characters were recorded using visual assessment, physical measurement, and RHS colour chart determinations.

Key Results

The evaluated genotypes showed variability across vegetative and flowering traits. Most genotypes had upright growth habit, medium stem girth, pinnate leaves, medium leaf length and width, flower heads above the foliage, upright flower-head attitude, medium flower-head length, decorative or pompon flower types, two ray-floret keels, pointed ray-floret apices, incurving ray-floret longitudinal axes, single-coloured ray florets, concave ray-floret cross sections, and elongated epicalyx shape. Most genotypes flowered early or mid-season and showed short vase life and elongated tubers. Six flower-colour groups were identified. Hierarchical cluster analysis separated the genotypes into distinct clusters.

Mechanistic Insight

Observed morphological variation was attributed to the interaction of genetic composition and environmental influence. Vase-life variation was described in the source as likely related to differences in carbohydrate accumulation among genotypes.

Practical Guidance

The characterized trait profiles provide reference data for identifying dahlia varieties in field and germplasm settings. The source notes that such profiles support improvement programs and help breeders and nurserymen seeking protection of new material under plant-variety protection systems.

Why This Source Matters

This source provides the most comprehensive morphological descriptor framework represented in this collection. Its fifty-one-trait protocol covers plant habit, leaf architecture, stem form, floral geometry, ray-floret characteristics, colour classification, phenology, vase life, and tuber shape. Its hierarchical cluster output demonstrates how genotypes can be separated systematically on morphological grounds.

The distinctness-uniformity-stability framework it employs represents the formal standard for cultivar characterization in plant-variety protection systems. The six flower-colour groups and trait-based cluster separations it documents provide a working model of what rigorous cultivar characterization looks like at the germplasm level.

Publication Type

Conference Proceedings Paper

Full Citation

Feng, L. J., Yuan, Z. H., Yin, Y. L., & Zhao, X. Q. (2010). Studies on the genetic diversity of phenotype characteristics for different dahlia populations. XXVIII International Horticultural Congress on Science and Horticulture for People (IHC2010): International Symposium on, 937, 411–418.

Study System

Eighty-four elite Dahlia cultivars grouped into four flower-type populations.

Experimental Context

Multi-year field evaluation of morphological traits across defined flower-type groups, assessing how much phenotypic variation occurs among populations versus within them.

Experimental Design

Seven phenotypic traits were measured across three years. Nested variance analysis was applied to partition trait variation into among-population and within-population components. Phenotypic differentiation coefficients were calculated for each trait to quantify the proportion of total variance attributable to among-population differences. Within-population correlation analysis was also conducted.

Key Results

Significant among-population differences were found for stem diameter, leaf length, leaf width, flower diameter, and pedicel length. Plant height and petiole length did not differ significantly among populations. The mean phenotypic differentiation coefficient across traits was 51.4 percent. Flower diameter showed the highest among-population differentiation at 75.4 percent. Leaf length and leaf width were positively correlated within all four populations.

Mechanistic Insight

Among-population phenotypic differentiation reflects the combined influence of genetic structure, environmental conditions, and the selection pressures applied in breeding programs that produce distinct flower-type populations. The high differentiation for flower diameter relative to vegetative traits suggests that selective emphasis on floral characters in dahlia breeding has produced more population-level divergence in floral geometry than in plant architecture.

Practical Guidance

Traits with higher phenotypic differentiation coefficients and greater variability among populations may serve as more useful selection targets in breeding programs, because they capture more of the genetic difference between classification groups. Leaf length and leaf width, which correlated within all populations, may be less independent as selection criteria than their separate measurement would suggest.

Why This Source Matters

This source documents how phenotypic variance is distributed among and within dahlia flower-type populations, providing a structured basis for evaluating how much of the observable difference between classification groups is captured in individual measurable traits. The finding that flower diameter accounts for the highest among-population differentiation supports its use as a primary descriptor in cultivar classification systems.

The nested variance design distinguishes this source from simple cultivar rankings by quantifying the population-level structure of phenotypic difference. It helps move the collection’s focus from named varieties to measurable trait patterns.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Ciobanu, I., & Cantor, M. (2016). Results regarding phenotypic correlations between morpho-decorative characteristics for some Dahlia variabilis cactus type cultivars. Lucrări Ştiinţifice Seria Horticultură, 59(1), 159–168.

Study System

Seven Dahlia variabilis cactus-type cultivars: ‘Kennemerland’, ‘Tsuki Yori No Sisha’, ‘Hayley Jane’, ‘Purple Gem’, ‘Star Favourite’, ‘Park Princess’, and ‘Friquolet’.

Experimental Context

Field-grown dahlia plants established from forced and unforced tuberous roots and assessed for morpho-decorative characteristics and phenotypic correlations.

Experimental Design

Fourteen experimental variants combined seven cultivars with two planting-material treatments: tuberous roots started earlier in a greenhouse before field planting, and tuberous roots planted directly in the field without prior forcing. Each variant used nine tuberous roots. Forcing began on 19 March 2015, and both treatments were planted in the field on 12 May 2015. Observations were repeated in 2016. Measurements included shoots per plant, flower-stem height, inflorescence diameter, inflorescences per plant, and ligulate florets per inflorescence. Results were assessed with the Duncan test at 5 percent probability, and phenotypic correlation coefficients were calculated among five trait pairs.

Key Results

Forced tuberous roots produced superior values for the analyzed morpho-decorative characteristics compared with unforced tuberous roots across cultivars. The highest shoot number was recorded for ‘Purple Gem’ from forced roots. The tallest flower stems and highest inflorescence number per plant were recorded for ‘Friquolet’ from forced roots. The largest inflorescence diameters were recorded for ‘Tsuki Yori No Sisha’ and ‘Star Favourite’ from forced roots. The highest number of ligulate florets per inflorescence was recorded for ‘Star Favourite’ from forced roots. More positive significant and distinctly significant correlation coefficients were recorded in the unforced variants than in the forced variants.

Mechanistic Insight

The source documents phenotypic associations among measured morpho-decorative traits but does not identify physiological or developmental mechanisms underlying the correlations. The direction and strength of correlations differed between forced and unforced treatments, indicating that planting-material preparation influences not only trait expression but also the relationships among traits.

Practical Guidance

Forcing tuberous roots before field planting increased measured morpho-decorative performance in the tested cactus-type cultivars. Cultivar responses differed across the five traits measured, meaning that no single cultivar outperformed all others across all characteristics. Evaluators comparing cactus-type cultivars for morpho-decorative purposes should specify whether planting material was forced or unforced, because the treatment affects both trait values and trait correlations.

Why This Source Matters

This source illustrates how phenotypic correlations among morpho-decorative traits can be measured within a defined dahlia flower-form class. Because the study works within the cactus type specifically, it keeps the descriptor space more homogeneous than a mixed-form cultivar trial would.

The comparison between forced and unforced planting material adds a practical production variable to the morpho-decorative evaluation. It demonstrates that the same cultivar can present differently depending on how planting material is prepared, a point directly relevant to any evaluation protocol that must produce consistent, comparable results.

Observing that two cultivars differ in a measured trait is not the same as knowing whether that difference is real, heritable, or useful for selection. Any field evaluation includes environmental variation, measurement error, and year-to-year fluctuation.

The sources in this cluster document the statistical tools researchers use to separate genetic signal from environmental noise: variance analysis, heritability estimates, genetic advance calculations, correlation coefficients, and path coefficient analysis. Together they address the question that underlies any serious cultivar comparison: how much of what we observe in a field trial reflects the cultivar itself?

Publication Type

Peer-Reviewed Journal Article

Full Citation

Devi, M. S., Seetharamu, G. K., Patil, B. C., Hanchinamani, C. N., Laxman Kukanoor, S. D., & Nishani, S. (2020). Assessment of genetic variability among dahlia (Dahlia variabilis L.) genotypes for productivity and quality traits. Journal of Pharmacognosy and Phytochemistry, 9(4), 3134–3137.

Study System

Thirty-two dahlia genotypes grown under open-field conditions in the northern dry zone of Karnataka, India.

Experimental Context

Genetic variability assessment for vegetative, flowering, yield, tuber, flower-quality, and vase-life traits in dahlia genotypes under field conditions.

Experimental Design

The experiment used a randomized block design with 32 genotypes, 60 cm by 40 cm spacing, and two replications, following recommended agro-techniques. Observations were recorded for eighteen vegetative and floral parameters. Mean, range, phenotypic coefficient of variation, genotypic coefficient of variation, broad-sense heritability, and genetic advance as percent of mean were calculated for each trait.

Key Results

Analysis of variance showed highly significant genotype differences for all eighteen traits studied. Phenotypic coefficient of variation was higher than genotypic coefficient of variation for all traits, indicating that environmental effects contributed to total observed variation in each case. High genotypic and phenotypic coefficients of variation were recorded for plant height at 90 days after planting, leaf area index, duration of flowering, number of flowers per plant, flower yield, tuber weight, number of tubers per plant, individual flower weight, flower diameter, petal length, and stalk length.

High heritability coupled with high genetic advance as percent of mean was recorded for plant height at 90 days after planting, leaf area index, plant spread, duration of crop, days to first flowering, days to 50 percent flowering, duration of flowering, number of flowers per plant, flower yield, tuber weight, number of tubers per plant, flower diameter, petal length, and stalk length.

Mechanistic Insight

The study interpreted significant genotype differences as evidence of genetic diversity among the evaluated genotypes. High heritability coupled with high genetic advance as percent of mean was interpreted as indicating additive gene action for those traits, meaning that the observed variation reflects differences that are passed on reliably under selection rather than being produced primarily by environmental fluctuation or non-additive genetic effects.

Practical Guidance

Traits showing high genotypic and phenotypic coefficients of variation together with high heritability and high genetic advance as percent of mean provide the most reliable basis for selection. The source identifies this combination as indicating that progress under selection is likely to be predictable, which is a useful criterion for deciding which traits to prioritize in a breeding or evaluation program.

Why This Source Matters

This source documents the range and heritability of eighteen traits across thirty-two dahlia genotypes, covering vegetative, flowering, yield, tuber, and flower-quality characteristics in a single field trial. The heritability and genetic advance data are particularly relevant for distinguishing which traits reflect stable genotypic differences from those more strongly shaped by the environment.

The breadth of traits covered, from leaf area index and plant spread through flower diameter, petal length, tuber weight, and vase life, makes this source a useful reference for understanding which dahlia characteristics carry the most consistent genetic signal across a field evaluation.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Beura, S., Maharana, T., & Jagadev, P. N. (1995). Genetic variability and character association in dahlia. Journal of Tropical Agriculture, 33, 20–22.

Study System

Fifteen genotypes of dahlia (Dahlia variabilis Desf.) grown in pot culture.

Experimental Context

Pot culture assessment of genetic variability, heritability, genetic advance, and phenotypic and genotypic correlation coefficients for vegetative, flowering, and flower-production traits.

Experimental Design

Plants were grown in 30 cm pots containing soil, compost, and sand in a 5:3:1 ratio, with one rooted cutting per pot and fifteen pots per replication across five replications. Fertilizer was applied after planting and again at first bud stage. Observations were recorded on ten traits from five randomly selected plants per replication: plant height, branches per plant, leaves per plant, internode length, stem diameter, leaf area, days to first flowering, diameter of first flower, flowering duration, and flowers per plant. Mean data were analyzed for variance, covariance, phenotypic and genotypic coefficients of variation, broad-sense heritability, expected genetic advance, and phenotypic and genotypic correlation coefficients.

Key Results

Total flowers per plant and its component traits showed significant genotype differences. High genotypic coefficient of variation with high heritability and high genetic advance was recorded for branches per plant, leaf area, and flowers per plant. Days to first flowering showed high heritability with low genotypic coefficient of variation and low genetic advance. Leaves per plant and flowering duration showed high heritability with moderate genotypic coefficient of variation and moderate genetic advance.

Genotypic and phenotypic associations were generally in the same direction, and genotypic correlation estimates were higher than phenotypic estimates throughout. Flowers per plant showed significant positive association with flowering duration and internode length, and significant negative association with days to first flowering.

Mechanistic Insight

The consistent finding that genotypic correlations exceeded phenotypic correlations indicates that the measured traits are inherently associated at the genetic level, with environmental effects partially masking those associations at the phenotypic level. The negative association between flowers per plant and days to first flowering suggests that early-flowering genotypes tend to produce more flowers across the season, while the positive association with branches per plant and flowering duration identifies the structural components most closely linked to total flower production.

Practical Guidance

Selection for improved flower production in dahlia should emphasize flowers per plant as the primary criterion, with supporting attention to branches per plant, flowering duration, and earliness of first flower. The source identifies these traits as the most genetically coherent cluster for improving flower yield under selection.

Why This Source Matters

This source provides a detailed early account of genotypic and phenotypic correlation structure among dahlia vegetative and flowering traits. The finding that genotypic correlations consistently exceeded phenotypic correlations is directly relevant to any evaluation system that relies on field observation, because it demonstrates that the observed relationships among traits are shaped partly by genetic associations that environmental effects partially obscure.

Understanding the trait correlation network matters for cultivar evaluation because changing one trait through selection predictably affects associated traits, whether or not those associations are immediately visible in field data.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Raghupathi, B., Sarkar, M. M., & Banerjee, S. (2019). Evaluation of genetic variability, correlation and path co-efficient analysis for cut flower attributing traits in medium decorative dahlia (Dahlia variabilis L.). Journal of Pharmacognosy and Phytochemistry, 8(1), 465–469.

Study System

Fifteen medium decorative dahlia cultivars evaluated under field conditions.

Experimental Context

Field-based assessment of genetic variability, heritability, genetic advance, correlation structure, and path coefficients for growth, cut flower, and tuber traits in medium decorative dahlia.

Experimental Design

A randomized block design with 15 cultivars and three replications was used. Growth, yield, and quality traits were recorded for each cultivar across replications. Phenotypic and genotypic coefficients of variation, broad-sense heritability, and genetic advance as percent of mean were calculated. Phenotypic and genotypic correlation coefficients were determined among trait pairs. Path coefficient analysis was then applied to partition the correlation between each trait and a primary outcome variable into a direct statistical effect of that trait and indirect statistical effects mediated through other measured traits.

Key Results

High heritability coupled with high genetic advance was recorded for several growth and yield traits, indicating additive gene action for those characteristics. Significant correlations were found among cut flower and tuber attributes. Path coefficient analysis showed that the simple correlations between certain traits and primary outcome variables were partly or substantially mediated through other traits in the measured set. In other words, some traits that appeared to have strong associations with an outcome exerted their influence indirectly through another measured trait rather than directly.

Mechanistic Insight

Additive gene effects govern the traits showing high heritability and high genetic advance, supporting the reliability of selection for those characteristics. The path analysis results indicate that trait relationships in dahlia are not always direct: a trait that correlates strongly with an outcome may do so because it is associated with a second trait that has the stronger direct statistical effect. Separating direct from indirect effects clarifies which traits exert independent influence on outcomes and which traits function mainly as correlated indicators.

Practical Guidance

Path coefficient analysis provides a more precise basis for selection decisions than correlation coefficients alone, because it identifies which traits have independent effects on a target outcome rather than effects that disappear when mediating traits are accounted for. The source identifies plant height, stalk traits, flowering time, and tuber weight as effective targets for selection based on the combined evidence of heritability and path structure.

Why This Source Matters

This source illustrates the methodological value of path coefficient analysis in dahlia cultivar evaluation. Where simple correlation shows which traits tend to move together, path analysis shows how much of that association is direct and how much is routed through other measured traits.

That distinction matters for any evaluation program trying to identify efficient selection targets. Selecting for a trait that acts mainly as an indirect indicator can be less efficient than selecting for the trait with the stronger direct statistical effect. Restricting the study system to the medium decorative class adds interpretive clarity, as KC-0115 does for cactus-type cultivars, by keeping the trait relationships within a more homogeneous phenotypic context.

Evaluating cultivars one trait at a time produces a long list of measurements without necessarily revealing which cultivars are genuinely similar to each other and which are genuinely distinct. Multivariate methods address this by analyzing many traits simultaneously, identifying which traits carry the most discriminating information, and grouping cultivars according to patterns of resemblance across the full trait set.

The sources in this cluster demonstrate three approaches to that problem: principal component analysis applied to germplasm panels, hierarchical clustering of seed-derived clonal populations, and grey relational analysis for multi-trait ornamental scoring.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Laguna C, A., Castillo G, F., Livera M, M., & López P, M. C. (1997). Phenotypic diversity in selected clones from a dahlia (Dahlia variabilis (Willd.) Desf.) seedling population. Revista Fitotecnia Mexicana, 20(1), 1–14.

Study System

One hundred Dahlia variabilis clones selected from a seed-derived population, field-evaluated in Chapingo, Mexico.

Experimental Context

Field evaluation of phenotypic diversity among clonal selections to assess the range of ornamental traits expressed in a seedling-derived population and to identify groupings useful for breeding and commercial advancement.

Experimental Design

A simple 10 by 10 lattice design with two replications was used. Eleven ornamental traits were measured. Analysis of variance was applied to test for differences among clones, followed by Tukey comparison tests. Principal component analysis and hierarchical cluster analysis were then applied to the full trait matrix to identify which traits accounted for the most variation and which clones grouped together across multiple traits simultaneously.

Key Results

Significant differences among clones were found for all traits measured. Hierarchical cluster analysis identified fourteen distinct groups among the hundred clones. The first four principal components explained 77.36 percent of total variation among clones. High variability was observed in petal number, capitula per plant, and flowering period.

Mechanistic Insight

The substantial genetic variability expressed in ornamental traits reflects the broad recombination potential within a seed-derived dahlia population. The finding that floral size, flowering timing, and plant architecture drove multivariate differentiation indicates that these trait complexes can accumulate genetic differences in a segregating population, rather than being constrained to vary together.

Practical Guidance

Multivariate analysis supports structured parent selection and recombination planning in a seedling population by revealing which clones are genuinely distinct across multiple traits and which cluster together in similar phenotypic space. Superior clones identified through principal component analysis and cluster analysis can advance to commercial evaluation with a more defensible basis than visual selection alone would provide.

Why This Source Matters

This source demonstrates the degree of phenotypic differentiation that can exist within a single seed-derived dahlia population. The fourteen-cluster outcome from one hundred clones shows that visual similarity can mask substantial multivariate divergence, and that principal component analysis and hierarchical clustering provide structure that single-trait ranking does not.

The combination of analysis of variance, principal component analysis, and cluster analysis applied to a seed-derived population makes this source a useful methodological reference for any evaluation of dahlia seedling or clonal material where the goal is to characterize diversity rather than simply rank performance.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Singh, S., Dhatt, K. K., & Bodla, P. K. (2023). Exploring genetic diversity of Dahlia (Dahlia variabilis Desf.) germplasm using multivariate statistics. Journal of Horticultural Sciences, 18(1), 67–76.

Study System

Dahlia variabilis genotypes evaluated under field conditions.

Experimental Context

Field evaluation of phenotypic and agronomic diversity across dahlia genotypes using multivariate statistical methods to identify the traits that most differentiate genotypes and to group genotypes by overall trait similarity.

Experimental Design

Field trait evaluation with principal component analysis applied to the full set of measured traits, followed by genotype differentiation analysis to identify which traits contributed most to variation among genotypes.

Key Results

Significant variability was observed across genotypes. Principal component analysis identified plant height, flower weight, stalk length, vase life, and flower number as the traits most strongly differentiating among genotypes in this study.

Mechanistic Insight

Phenotypic divergence among genotypes across these traits provides information about breeding potential, because genotypes that differ across multiple independently differentiating traits represent contrasting parents with the potential to produce diverse offspring. The traits identified through principal component analysis as highest-loading are those that capture the most information per measurement across the genotype panel.

Practical Guidance

Multivariate trait data can be used to identify contrasting parents and to prioritize breeding lines for specific goals involving flower yield, vase life, and plant architecture. Principal component analysis based trait identification provides a more efficient basis for parent selection than evaluating all traits with equal weight.

Why This Source Matters

This source demonstrates the use of principal component analysis to identify which traits most strongly differentiate dahlia genotypes, offering a structured approach to germplasm characterization that goes beyond single-trait ranking. The five traits it identifies as key differentiators, plant height, flower weight, stalk length, vase life, and flower number, span vegetative architecture, floral production, and postharvest quality.

That range matters because cultivar evaluation rarely depends on only one trait. Multivariate analysis helps reveal which dimensions of variation carry the most useful information in a genotype panel.

Publication Type

Journal Article

Full Citation

Li, R. N. (2019). Landscape appreciation of fifty dahlia species in Xi’an area. Hubei Agricultural Sciences, 58(3), 68–72.

Study System

Fifty cultivars of Dahlia pinnata Cav. field-grown in the Xi’an region of China. The word “species” in the source title reflects translation convention; the study evaluates cultivated dahlia rather than wild species.

Experimental Context

Regional landscape evaluation of cultivated dahlia in Xi’an, assessing ornamental performance across multiple traits simultaneously under local growing conditions.

Experimental Design

Fifty cultivars were grown in field conditions and evaluated across multiple ornamental traits. Scores across traits were combined using grey relational analysis, a method that converts multi-trait data into composite similarity coefficients relative to an ideal reference profile, allowing cultivars to be ranked by overall ornamental value rather than by any single characteristic.

Key Results

Substantial variation in ornamental performance was identified among the fifty cultivars. Grey relational analysis identified cultivars with the highest composite ornamental value. Integrated assessment of plant architecture and flowering traits determined overall ornamental ranking, with no single trait dominating the outcome.

Mechanistic Insight

Grey relational analysis weights each measured trait relative to a defined ideal and combines the weighted scores into a composite index. This allows differences across multiple traits to contribute proportionally to the final ranking, rather than letting any one strong or weak trait determine the outcome for a given cultivar.

Practical Guidance

Multi-trait composite scoring provides a more defensible basis for cultivar selection decisions in landscape or ornamental use than ranking by single traits such as flower size or stem height alone. The grey relational analysis framework can be adapted to evaluation contexts where the goal is to select based on balanced performance across several desirable characteristics.

Why This Source Matters

This source demonstrates the use of grey relational analysis as a structured multi-trait evaluation method for cultivated dahlia. It provides a concrete example of how ornamental performance can be quantified and compared across cultivars in a regional field setting.

The composite-scoring approach it employs is directly relevant to evaluation problems where no single trait fully captures the quality of interest. That is common in ornamental plant assessment, where plant habit, flowering display, bloom traits, and landscape effect all matter at the same time.

A cultivar that performs well in one location, season, or management system does not necessarily perform the same way elsewhere. This is not a failure of the research. It is the expected outcome of growing plants whose trait expression responds to temperature accumulation, planting date, and local environment.

The sources in this final cluster document two aspects of that problem: direct evidence that cultivar and location effects interact in measurable ways to shape development, and molecular evidence that horticultural classification categories themselves do not map cleanly onto the genetic structure of the cultivar pool.

Together they make the case for treating regional trial results as evidence about how traits are expressed under defined conditions rather than as transferable rankings.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Fernandes, M. E. S., Roso, T. P., Ferronato, L., Freitas, C. P. D. O. D., Tomiozzo, R., Uhlmann, L. O., Zanon, A. J., & Streck, N. A. (2023). Determining the phyllochron and final leaf pair number in on-farm cut dahlia cultivars. Ornamental Horticulture, 29(2), 299–312.

https://scholar.google.com/scholar?q=Determining+the+phyllochron+and+final+leaf+pair+number+in+on-farm+cut+dahlia+cultivars+Fernandes+Roso+Streck+2023

Study System

Cut dahlia cultivars grown on farms across six locations in Rio Grande do Sul State, southern Brazil, under humid subtropical conditions.

Experimental Context

Three on-farm experiments were conducted during 2021 and 2022. Experiments included two to eight cut dahlia cultivars depending on tuber availability at each farm. The study aimed to quantify leaf appearance rate and final leaf pair number using thermal time accumulation across multiple cultivars and locations.

Experimental Design

Tubers were planted in beds at 0.40 m by 0.40 m spacing. Ten plants per cultivar were tagged after emergence. Unfolded leaf pairs on the main shoot were counted once or twice weekly until the final leaf pair unfolded. Accumulated thermal time from emergence was calculated using established base, optimum, and upper base temperatures for dahlia leaf pair unfolding. Leaf pair number was linearly regressed against accumulated thermal time for each plant, and phyllochron was estimated as the inverse of the leaf pair unfolding rate. Phyllochron and final leaf pair number were analyzed across experiments, locations, and cultivars using nonparametric statistical tests.

Key Results

Leaf pair number showed a linear relationship with accumulated thermal time, with R-squared values greater than 0.9 in all locations and experiments. Phyllochron ranged from 27.2 to 106.4 degrees Celsius day per leaf pair across the study. Final leaf pair number ranged from 6 to 15 leaf pairs per stem. The median final leaf pair number was 11 leaf pairs per stem in all three experiments. Phyllochron and final leaf pair number did not differ significantly among experiments. Location effect was significant for phyllochron but not for final leaf pair number. Cultivar effect was significant for both phyllochron and final leaf pair number.

Mechanistic Insight

Leaf appearance in the evaluated dahlia cultivars was primarily associated with accumulated thermal time rather than calendar time, meaning that developmental rate is driven by temperature sum rather than elapsed days. Final leaf pair number, while still cultivar-associated, was less affected by location than phyllochron, indicating that the two traits have different sensitivities to local environmental conditions.

Practical Guidance

Cultivar differences in phyllochron and final leaf pair number need to be taken into account when comparing vegetative development among cut dahlia cultivars evaluated at different locations or in different seasons. The same cultivar will reach equivalent developmental milestones at different calendar dates depending on accumulated thermal time. Comparison studies conducted across sites or years therefore need to account for temperature environment rather than treating planting date as a sufficient descriptor of growing conditions.

Why This Source Matters

This source establishes that leaf appearance rate is primarily driven by thermal time accumulation and that cultivar differences in that rate are real and significant. It also shows that location modifies the rate in ways that calendar-based evaluation would miss.

The finding that phyllochron varies significantly by location while final leaf pair number does not identifies two traits with different degrees of environmental sensitivity within the same cultivar. That distinction is directly relevant to any evaluation protocol designed to produce results that can be compared across growing environments.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Fernandes, M. E. S., Freitas, C. P. D. O. D., Tomiozzo, R., Simon, M. L., Silva, L. G. O. D., Uhlmann, L. O., Zanon, A. J., & Streck, N. A. (2024). Understanding the dynamics of vegetative and reproductive development in on-farm cut dahlia. Ornamental Horticulture, 30, e242754.

https://scholar.google.com/scholar?q=Understanding+the+dynamics+of+vegetative+and+reproductive+development+in+on-farm+cut+dahlia+Fernandes+Streck+2024

Study System

Cut dahlia cultivars evaluated across multiple on-farm locations and growing seasons.

Experimental Context

Multi-location phenological monitoring of cut dahlia to characterize how vegetative and reproductive development unfold under on-farm growing conditions, and to identify the developmental factors that most strongly determine overall cycle duration.

Experimental Design

Phenological development was monitored at multiple sites across growing seasons using thermal time accumulation and phyllochron analysis to characterize vegetative and reproductive phase dynamics.

Key Results

Vegetative phase duration was identified as the primary determinant of overall developmental cycle length. Strong cultivar and planting date effects were found, with both factors shaping the timing and duration of development in ways that interacted with growing environment.

Mechanistic Insight

Leaf appearance rate drives flowering timing more directly than final leaf pair number, meaning that how quickly a cultivar produces successive leaf pairs under a given temperature regime, rather than how many it ultimately produces, is the more critical factor in determining when reproductive development begins. This finding reinforces that thermal environment, cultivar, and planting date interact in determining developmental outcomes.

Practical Guidance

Cultivar selection and planting date choices interact with local temperature environment to determine developmental cycle duration and harvest timing. Calendar-based planning that ignores thermal accumulation will produce inconsistent results across locations and seasons, because the same calendar interval represents different amounts of thermal time depending on the growing environment.

Why This Source Matters

This source extends the thermal-time framework established in KC-0094 to the reproductive phase of dahlia development, showing that vegetative phase duration, driven by phyllochron, sets the conditions for reproductive development rather than simply preceding it. The on-farm multi-location design and the documented interaction between cultivar, planting date, and environment make this source a direct argument against treating cultivar evaluation results from one location or season as straightforwardly applicable elsewhere.

The fields available for this Knowledge Card are less detailed than those for KC-0094, but the companion-paper relationship is still useful. Together, the two sources show how a specific measured trait, leaf appearance rate, can connect cultivar identity, thermal environment, planting date, and flowering schedule.

Publication Type

Peer-Reviewed Journal Article

Full Citation

Wegner, H., & Debener, T. (2008). Novel breeding strategies for ornamental dahlias II: Molecular analyses of genetic distances between dahlia cultivars and wild species. European Journal of Horticultural Science, 73(3), 97–103.

https://scholar.google.com/scholar?q=Novel+breeding+strategies+for+ornamental+dahlias+II+molecular+analyses+genetic+distances+Wegner+Debener+2008

Study System

Nineteen Dahlia cultivars, three wild Dahlia genotypes, and one wild hybrid.

Experimental Context

Assessment of genome-wide genetic diversity among cultivated dahlia and wild Dahlia species using molecular markers, with the goal of understanding how genetic distances relate to horticultural classification and to inform breeding parent selection.

Experimental Design

AFLP profiling was conducted using ten primer combinations, scoring 1,432 markers across all accessions. Genetic similarity was calculated using the Jaccard coefficient. A UPGMA dendrogram was constructed from the similarity matrix and evaluated with bootstrap analysis.

Key Results

Polymorphism across scored markers was 76.5 percent. Cultivars clustered separately from wild species in the dendrogram. Average genetic similarity among cultivars ranged from 0.68 to 0.77. Genetic clustering based on AFLP markers did not correlate with horticultural classification categories based on flower form.

Mechanistic Insight

Genome-wide AFLP markers capture genetic differences distributed across the entire genome rather than concentrated in the visible traits that define horticultural classification categories. The absence of correlation between molecular clustering and form-based classification indicates that cultivars placed in the same horticultural class on the basis of flower type may be quite different in their overall genetic backgrounds, and that form similarity reflects convergence in a small number of traits rather than broad genetic relatedness.

Practical Guidance

Breeding parent selection based on flower-form classification alone may miss meaningful genetic distance between candidates. Molecular distance data provide a complementary basis for identifying parents that differ broadly at the genome level, which can contribute to greater offspring diversity than selecting parents that differ only in the trait categories captured by horticultural classification.

Why This Source Matters

This source provides direct molecular evidence that horticultural classification categories do not reliably reflect genome-wide genetic relationships in dahlia. The absence of correlation between AFLP-based genetic clustering and flower-form classification is a concrete demonstration that visible cultivar categories should not be treated as proxies for genetic identity or breeding distance.

In the context of this collection, this finding closes the framing opened at the beginning: descriptor systems and classification categories are valuable and necessary tools, but they describe a cultivar’s observable traits and not its full genetic background.

The research in this collection converges on a straightforward point that is easy to lose in the details of any individual study. Descriptor systems, variance statistics, heritability estimates, multivariate clustering, and thermal-time models are all ways of making dahlia traits more comparable and more useful for decision-making.

They work. They produce real information about how cultivars differ, which traits are heritable, and which groupings are genuinely distinct. But they each describe the cultivar as it expressed itself under the conditions of that study, at that location, in that season, with those methods.

The molecular evidence in the final card adds a further layer: even the flower-form categories used to classify cultivars do not capture the full genetic distances among them.

None of this makes cultivar evaluation less useful. It makes the results more interpretable. A grower or breeder reading a regional trial or a morphological characterization study is looking at a measured snapshot of cultivar expression, not a permanent ranking. The methods documented here are what allow that snapshot to be taken rigorously and compared honestly with snapshots taken elsewhere.

The Knowledge Card summaries in this collection were developed from the Dahlia Doctor research archive and checked against available source records during editorial preparation. AI tools assisted with retrieval, formatting, comparison of candidate Knowledge Cards, and assembly of the collection. All curatorial decisions, including source selection, topic organization, citation corrections, interpretation, and final editorial framing, were made by the author.