Nutritional and Fiber Quality Assessment of Native Greek Dactylis glomerata Populations

Abstract

Dactylis glomerata, a perennial forage grass widely distributed in Mediterranean areas, is recognized for its adaptability and nutritional quality. This study aimed to assess the chemical composition and fiber components of ten natural populations of Dactylis glomerata in order to characterize genetic variability in nutritional and fiber traits among populations. Seeds of all populations were established in a randomized complete block design with four replicates and cultivated for two consecutive years. Forage was collected at the boot stage, and analyses were conducted for crude protein, ash, crude fiber, neutral and acid detergent fibers, acid detergent lignin, hemicellulose, cellulose, digestible dry matter, dry matter intake, and relative feed value. Combined ANOVA indicated that genotypic effects were highly significant for all traits (p ≤ 0.001), with additional significant contributions from environmental and genotype × environment interactions. Crude protein ranged from 11.74% to 14.98%, neutral detergent fiber from 56.31% to 58.43%, and relative feed value from 100.1 to 106.4 among populations. Stability index analysis identified Kefalopotamos and Filyra as the most environmentally stable populations, whereas Kori and Xyloparoiko exhibited relatively higher values in selected forage quality traits. Broad-sense heritability values were high for the majority of traits (H² between 93.3% and 99.9%, except for hemicellulose), suggesting a strong genetic influence. Correlation analysis also revealed inverse relationships between protein content and fiber fractions and positive relationships with digestibility-related indices. Multivariate analyses revealed a clear separation between nutritional quality traits and structural fiber components, indicating consistent differentiation among populations. Overall, these results highlight the potential of local Dactylis glomerata populations as genetic resources for further evaluation in breeding and conservation programs under Mediterranean conditions.

1. Introduction

Dactylis glomerata L., commonly known as orchardgrass, is a long-lived grass species with wide distribution, recognized for its adaptability and forage quality [1,2]. It is commonly used in pastures and mixed grass-legume systems [3,4]. Orchardgrass provides high-quality forage with elevated crude protein content, moderate fiber levels, and good digestibility, making it a preferred species for ruminant nutrition [5,6,7,8]. Under specific pasture management conditions, grazing animals fed D. glomerata can exhibit improved liveweight gain and milk production compared to other grasses [9,10].

The nutritive value of D. glomerata is influenced by environmental conditions, plant maturity, and morphological traits such as leaf-to-stem ratio and tillering capacity. Key chemical and fiber constituents—including crude protein (CP), crude fiber (CF), neutral and acid detergent fibers (NDF and ADF), acid detergent lignin (ADL), cellulose, and hemicellulose—differ among genotypes and are also shaped by management practices, harvest stage, and environmental stresses [2,8,11,12,13]. Harvest at the vegetative stage produces higher CP and digestible dry matter (DDM), while fiber fractions increase with plant maturity [2]. Genetic variation among populations influences DDM, CP, and water-soluble carbohydrate content, indicating the existence of exploitable phenotypic diversity for forage quality traits [8,13]. Multi-environment testing is essential to evaluate the stability of these traits [12]. Given this variability, assessing local Greek populations is essential to identify genotypes with differential expression of nutritional and fiber traits under Mediterranean conditions.

In Greece, studies on orchardgrass have primarily focused on genetic, morphological, and agronomic traits of natural populations, whereas detailed information on how these traits vary under different environmental conditions is limited [14,15]. Previous research has investigated the effects of factors such as shade, water availability, and temperature on plant performance, including growth and dry matter production [14,15]. Research across Europe and the Mediterranean has shown high genetic variability and geographic structuring in D. glomerata populations, despite limited ploidy variation [11,13,15,16,17,18,19]. Genetic and phylogenetic analyses of subspecies reveal substantial diversity, emphasizing the need to consider subspecies variation in breeding and conservation programs [20,21].

Genetic variation influences several forage quality attributes, including growth traits and compositional parameters [13], while growth, persistence, and post-drought recovery differ among genotypes [11]. Maintaining consistent performance of these traits across different environments is important for pasture management, as understanding genotype × environment (G × E) interactions helps breeders identify genotypes that perform reliably under diverse conditions [22,23]. For instance, Greveniotis et al. [12] reported significant G × E interactions for digestible dry matter, intake of dry matter, and relative feed value, demonstrating how environmental factors shape genetic potential.

Standard analytical methods—including determination of CP, ash, CF, NDF, ADF, ADL, and derived indices such as DDM, DMI, and RFV—allow reliable comparisons across populations and provide insights into both nutritional and fiber traits, as these methods are widely used in forage quality assessment [24,25,26].

In Mediterranean regions, forage yields are often reduced due to drought, heat, and seasonal variations [12,27]. Under such conditions, the use of locally adapted germplasm is particularly important, as native populations are more likely to possess traits related to stress tolerance and stable performance [28,29]. Therefore, the evaluation of indigenous Dactylis glomerata populations is essential for identifying genotypes with improved adaptability and forage quality suitable for Mediterranean grazing systems.

Based on the observed genetic variability and potential genotype × environment (G × E) interactions among Dactylis glomerata populations, we hypothesized that local populations from different natural sites in the Trikala region would exhibit significant differences in chemical composition, fiber fractions, and overall forage quality, and that variation in trait expression would reflect underlying genetic diversity among populations. The present research focused on analyzing the chemical composition, fiber fractions, and forage quality of Dactylis glomerata populations from ten natural sites in the Trikala region, Greece, grown in experimental field plots under controlled management conditions, in order to assess phenotypic variation, genetic variability, and trait associations in forage quality characteristics among native populations.

2. Materials and Methods

2.1. Collection and Sampling of Seeds

Seeds of Dactylis glomerata were collected from natural populations in the Trikala region, Greece, during 2023, representing local autochthonous genotypes. Collection was conducted under authorization from the Ministry of Environment and Energy, according to Flora Research Permit Protocol No. ΥΠΕΝ/ΔΠΔ/129149/7625/10-01-2023 (ADA 6Ρ9Γ4653Π8-ΖΦΚ).

Seeds were harvested from multiple individuals per population to ensure genetic diversity while minimizing impact on natural populations. The primary aim of the collection was ex situ conservation and evaluation of autochthonous Greek populations with agronomic interest.

Collected seeds were stored under dry and cool conditions until sowing for the experiment.

Table 1. Collection sites and dates of Dactylis glomerata populations in the Trikala region.

Genus	Species	Material Ex Situ	Collection Date	Prefecture	Location
Dactylis	D. glomerata	Seeds	22 September 2023	Trikala	Palaiochori, former Municipality of Pialeia (G1)
Dactylis	D. glomerata	Seeds	25 October 2023	Trikala	Gorgogyri (G2)
Dactylis	D. glomerata	Seeds	31 October 2023	Trikala	Xyloparoiko (G3)
Dactylis	D. glomerata	Seeds	1 November 2023	Trikala	Prinos (G4)
Dactylis	D. glomerata	Seeds	9 November 2023	Trikala	Pialeia (G5)
Dactylis	D. glomerata	Seeds	14 November 2023	Trikala	Prodromos (G6)
Dactylis	D. glomerata	Seeds	16 November 2023	Trikala	Filyra (G7)
Dactylis	D. glomerata	Seeds	21 November 2023	Trikala	Kefalopotamos (G8)
Dactylis	D. glomerata	Seeds	23 November 2023	Trikala	Kori (G9)
Dactylis	D. glomerata	Seeds	29 November 2023	Trikala	Kato Rachi (Neromylos) (G10)

summarizes the populations, collection dates, and locations of the harvested seeds. Although collections were conducted over a period from September to November, seeds were harvested at physiological maturity in each location, ensuring a comparable developmental stage among populations.

2.2. Experimental Design

2.2.1. Field Trials

The field experiment was initiated in late November 2023 in the Trikala region (39°55′ N, 21°46′ E, 120 m above sea level). Two successive growing periods were included to capture a wider range of environmental conditions, facilitating the evaluation of genotype performance and stability across environments.

A Randomized Complete Block Design (RCBD) with four replications was employed. Each plot consisted of four rows, 2 m long, spaced 25 cm apart. Data were collected over two experimental years and analyzed both within individual years and as a combined dataset for statistical evaluation.

2.2.2. Soil Characteristics

The experimental soil was loamy, composed of 40% sand, 33% silt, and 27% clay. Key soil properties were: N–NO₃ 5.7 mg kg⁻¹, P–Olsen 14.4 mg kg⁻¹, K 175 mg kg⁻¹, organic matter 1.8%, and CaCO₃ 8.48%.

2.2.3. Crop Management

Plots were sown at a rate of 15 kg ha⁻¹, reflecting a typical plant density for Dactylis glomerata under field conditions, as suggested by previous studies [30]. Uniform and satisfactory stand establishment was achieved across all plots, ensuring comparable plant density and minimizing differences in plant competition among genotypes. Annual fertilization consisted of 150 kg N ha⁻¹ and 80 kg P₂O₅ ha⁻¹. In the first year, the entire phosphorus dose and 50 kg N ha⁻¹ were applied before planting, with the remaining 100 kg N ha⁻¹ split into four top-dressings of 25 kg N ha⁻¹ each during the growing season, following rainfall or irrigation. In the second year, the same fertilization rates and top-dressing schedule were followed. All fertilizers were applied on the soil surface without incorporation. Weed and pest management were applied uniformly across all plots following local best practices. Irrigation was applied using a conventional sprinkler system, based on crop water requirements and prevailing weather conditions, in order to avoid water stress and maintain uniform growth across plots. Irrigation management was consistent across all treatments, ensuring uniform soil moisture conditions throughout the experimental period.

2.2.4. Climatic Data

Climatic conditions during the experimental periods were recorded to provide context for plant growth and to facilitate the interpretation of potential environmental effects on the evaluated traits. Average monthly minimum and maximum temperatures (°C) and total monthly precipitation (mm) for the experimental periods are presented in Figure 1.

2.3. Harvesting and Analyses

Plants were harvested at the boot stage (just prior to flowering) during spring 2024 and 2025 to balance high crude protein content with moderate fiber levels, ensuring optimal plant material for nutritional and fiber analyses. For each plot, the entire area was cut to 5 cm above ground; to reduce edge effects, only the central rows were collected and used as the experimental unit for statistical analyses of all chemical and fiber assessments. Analyses were carried out at the Laboratory of Animal Technology, University of Thessaly.

Dried samples were prepared according to standard AOAC procedures [31,32], oven-dried at controlled temperature conditions (IKA 125, IKA-Werke GmbH & Co. KG, Staufen im Breisgau, Germany), and ground through a 1 mm sieve (Retsch ZM200, Retsch GmbH, Haan, Germany) to ensure uniformity. Crude Protein (CP) was estimated from nitrogen content (N × 6.25) using the Kjeldahl method (AOAC 990.03) with a Gerhardt Kjeldatherm KT20s digestion block and Gerhardt Vapodest 300 distillation unit. Ash content was determined by incineration at 600 °C for 2 h (AOAC 942.05) and expressed as a percentage of dry matter. Crude Fiber (CF), representing cellulose and hemicellulose resistant to digestion, was measured through sequential acid and alkaline digestion (AOAC 978.10), with ash correction.

Acid Detergent Fiber (ADF), Neutral Detergent Fiber (NDF), and Acid Detergent Lignin (ADL) were analyzed following Van Soest et al. [25] using a Velp FIWE Fiber Analyzer (VELP Scientifica, Usmate, Italy). Cellulose and hemicellulose contents were calculated as ADF minus ADL and NDF minus ADF, respectively [33].

Digestible Dry Matter (DDM, %) was calculated using the formula DDM = 88.9 − 0.779 × ADF (%), and Dry Matter Intake (DMI, % of body weight) as DMI = 120 ÷ NDF (%), in accordance with standard forage evaluation protocols [34,35]. Relative Feed Value (RFV) was subsequently derived as RFV = (DDM × DMI)/1.29 [34]. RFV categories provide a practical guide for livestock feeding: Class I (>151) for high-producing dairy cows, Class II (125–151) for good-quality dairy cows and young heifers, Class III (103–124) for beef cattle and older heifers, Class IV (87–102) for beef cattle and non-lactating cows, and Class V (75–86) for dry cows [34].

All measurements were conducted in triplicate per plot, and mean values were used for statistical analysis. Samples from each experimental year (2024 and 2025) were analyzed separately. For the combined analysis, mean values per plot and year were used as input in the statistical model, allowing the assessment of both environmental effects and genotype × environment interactions. This approach minimizes environmental and procedural variation, allowing a clearer assessment of genetic differences among populations.

2.4. Statistical Analyses and Trait Stability

All statistical analyses were carried out to explore variation and relationships among the evaluated forage quality traits. A combined ANOVA was performed for each trait, treating genotypes as fixed effects [36], using IBM SPSS v.31 (IBM Corp., Armonk, NY, USA). Mean comparisons were conducted using Duncan’s multiple range test at p < 0.05, implemented in MSTAT-C (Michigan State University, East Lansing, MI, USA, v.2.10). Given the relatively small numerical differences observed among population means, results of the mean comparison test should be interpreted with caution, as statistically significant differences may not always reflect biologically meaningful variation.

To assess associations among traits, Pearson correlation coefficients were calculated and visualized in JMP Pro 18 (SAS Institute Inc., Cary, NC, USA), with statistical significance evaluated at p < 0.05. To further explore overall patterns and relationships among traits and populations, Principal Component Analysis (PCA), network ordination, and two-way Hierarchical Cluster Analysis (HCA) were performed. For the HCA, data were standardized to account for differing scales, and Ward’s minimum variance method was applied using Euclidean distance as the proximity metric. Clustering relationships were visualized via a dual-axis heat map and dendrograms presented on a Distance Scale, complemented by a constellation plot to illustrate the structural arrangement and relative distances between clusters.

The Stability Index (SI) for each trait was calculated following Fasoula [37] as:

$$\mathrm{S}\mathrm{I}= (\bar{x}/s{)}^2$$

where $\bar{x}$ represents the mean value of the trait and $s$ is the standard deviation, with higher values reflecting greater stability of the trait across different environments.

The Stability Index (SI) was then used to evaluate genotype stability based on the relationship between mean performance and variability, providing a simple and effective criterion for identifying stable genotypes, particularly in experiments with a limited number of environments.

Variance components were derived from the ANOVA mean squares according to McIntosh [38] and used to estimate broad-sense heritability (H²) following Johnson et al. [39] and Hanson et al. [40]. The phenotypic (PCV) and genotypic (GCV) coefficients of variation were calculated as described by Singh and Chaudhary [41] to assess the relative contribution of genetic and environmental factors to trait variability.

$${\mathrm{H}}^2={\displaystyle \frac{{\sigma }_g^2}{{\sigma }_g^2+\frac{{\sigma }_{gxe}^2}{e}+\frac{{\sigma }_{re}^2}{rxe}}}$$

$$\mathrm{GCV}(\%)={\displaystyle \frac{\sqrt{{\sigma }_g^2}}{\overline{x}}}\times 100$$

$$\mathrm{PCV}(\%)={\displaystyle \frac{\sqrt{{\sigma }_p^2}}{\overline{x}}}\times 100$$

For these calculations, genotypic variance, phenotypic variance, genotype × environment interaction variance, residual error variance, number of replications, number of environments, and overall mean were considered as ${\sigma }_g^2$, ${\sigma }_p^2$, ${\sigma }_{gxe}^2$, ${\sigma }_{re}^2$, r, e, and $\overline{x}$, respectively.

3. Results

Forage quality traits of ten Dactylis glomerata populations were evaluated over two consecutive years under controlled conditions to determine the effects of genotype, environment, and their interaction on chemical composition, fiber fractions, and derived indices (DDM, DMI, RFV).

3.1. Combined ANOVA

The combined ANOVA for Dactylis glomerata populations (

Table 2. Combined ANOVA for Dactylis glomerata populations. Traits include crude protein (CP), ash, crude fiber (CF), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), hemicellulose, cellulose, digestible dry matter (DDM) (all expressed as %), dry matter intake (DMI, % of body weight), and relative feed value (RFV). Significance of population, environment, and their interaction is shown for each trait.

Source of Variation	CP	Ash	CF	NDF	ADF	ADL	Hemicellulose	Cellulose	DDM	DMI (% BW)	RFV
	m.s.	m.s.	m.s.	m.s.	m.s.	m.s.	m.s.	m.s.	m.s.	m.s.	m.s.
Environment (E)	1.200 ***	0.553 ***	0.008 ***	0.450 ***	0.412 ***	0.012 ***	1.723 ***	0.281 ***	0.250 ***	0.001 ***	0.157 ***
REPS/ Environments	0.001 ns	0.001 ns	0.0005 ns	0.0002 ns	0.0002 ns	0.0002 ns	0.0003 ns	0.0005 ns	0.0002 ns	0.000001 ns	0.0003 ns
Genotype (G)	8.760 ***	3.792 ***	6.368 ***	3.939 ***	6.691 ***	0.195 ***	2.756 ***	6.651 ***	4.061 ***	0.005 ***	41.498 ***
Environment × Genotype (E × G)	0.003 ***	0.002 ***	0.297 ***	0.243 ***	0.238 ***	0.008 ***	0.960 ***	0.246 ***	0.144 ***	0.0003 ***	0.076 ***
Error	0.001	0.0004	0.0003	0.0005	0.0004	0.0003	0.001	0.001	0.0002	0.000001	0.003

Probability levels: *** p ≤ 0.001; ns—not significant; m.s.: mean squares.

) showed that genotypic effects were significant for all evaluated traits (p ≤ 0.001), including crude protein (CP), crude fiber (CF), and relative feed value (RFV). Environmental effects were also significant for all traits, although generally lower than those for genotypes, indicating that both genetic variation and environmental factors, including genotype × environment interaction, play an important role in determining forage quality. The genotype × environment interaction was significant for all traits, reflecting differential responses of genotypes across environments. Replications within environments were not significant for any trait.

These results highlight a strong genetic influence on forage quality traits, while also indicating substantial contributions of environmental factors and genotype × environment interactions. Replications nested within environments were not significant, indicating low experimental error and good field uniformity.

3.2. Stability of Forage Quality Traits

The Stability Index (SI) analysis for Dactylis glomerata populations (

Table 3. Stability Index (SI) values for Dactylis glomerata populations, including crude protein (CP), ash, crude fiber (CF), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), hemicellulose, cellulose, and digestible dry matter (DDM) (all expressed as %), dry matter intake (DMI, % of body weight), and relative feed value (RFV). Higher SI values indicate that a population maintains more consistent trait performance under varying environmental conditions.

Population	CP	Ash	CF	NDF	ADF	ADL	Hemicellulose	Cellulose	DDM	DMI (% bw)	RFV
Palaiochori	8318	4658	15,187	88,830	22,799	5341	3727	28,744	82,002	86,244	94,365
Gorgogyri	7741	4382	15,868	81,060	30,523	6405	4402	17,287	87,945	80,758	89,280
Xyloparoiko	13,225	6402	17,136	87,645	25,486	5129	4115	31,795	89,473	86,519	91,106
Prinos	8035	6157	19,397	93,979	30,720	5990	4061	17,642	89,789	93,169	92,392
Pialeia	9499	6205	25,220	80,107	33,640	6618	4308	18,854	89,954	76,403	89,170
Prodromos	11,312	6635	20,490	82,765	28,606	7230	3938	34,217	91,426	79,906	90,322
Filyra	9326	7147	19,985	91,332	28,946	6372	4510	15,436	94,568	91,687	91,031
Kefalopotamos	13,094	7532	18,672	93,483	30,192	8611	3937	34,380	95,163	92,638	93,898
Kori	10,745	9310	23,734	83,290	31,137	7608	4838	30,341	97,725	82,377	89,045
Kato Rachi	11,626	7484	25,615	86,508	28,855	8200	4114	33,517	92,104	83,684	90,861

) revealed considerable variation in environmental stability among the evaluated populations. Certain populations, such as Kefalopotamos, and Filyra, exhibited high SI values for key nutritional traits—including digestible dry matter (DDM), dry matter intake (DMI), and relative feed value (RFV)— indicating high environmental stability (i.e., consistent performance across environments). In contrast, populations like Gorgogyri and Pialeia showed lower stability for several traits, greater sensitivity to environmental variation.

Similarly, crude protein (CP) and neutral detergent fiber (NDF) stability varied across populations, highlighting differences in how genotypes maintain quality traits under variable environmental conditions. Overall, these results suggest that some populations combine both superior forage quality and environmental stability, making them potential candidates for further evaluation in breeding programs; however, selection decisions should also consider additional agronomic traits such as dry matter yield and persistence.

3.3. Descriptive Statistics

The genetic parameter estimates for the Dactylis glomerata populations indicated strong genetic control over most of the assessed forage quality traits (

Table 4. Genetic parameter estimates for Dactylis glomerata populations, including crude protein (CP), ash, crude fiber (CF), acid detergent fiber (ADF), neutral detergent fiber (NDF), acid detergent lignin (ADL), hemicellulose, cellulose, and digestible dry matter (DDM) (all expressed as %), dry matter intake (DMI, % of body weight), and relative feed value (RFV). The table presents minimum (Min), maximum (Max), mean, standard deviation (SD), genotypic variance (${\sigma }_g^2$), phenotypic variance (${\sigma }_p^2$), genotypic coefficient of variation (GCV %), phenotypic coefficient of variation (PCV %), and broad-sense heritability (H²%). These parameters reflect genetic contributions to trait variation and may be useful to identify populations with superior nutritional and fiber qualities suitable for breeding programs.

Traits	Min.	Max.	Mean	SD	${\mathit{\sigma }}_{\mathit{g}}^2$	${\mathit{\sigma }}_{\mathit{p}}^2$	GCV (%)	PCV (%)	H2 (%)
CP	11.61	14.97	13.275	1.007	1.0946	1.095	7.8813	7.8827	99.97
Ash	6.16	8.37	7.259	0.663	0.4738	0.4740	9.4818	9.4843	99.95
CF	26.11	28.79	27.475	0.872	0.7589	0.7960	3.1707	3.2473	95.34
NDF	56.11	58.63	57.418	0.695	0.4620	0.4924	1.1838	1.2221	93.83
ADF	30.72	33.89	32.502	0.892	0.8066	0.8364	2.7633	2.8138	96.44
ADL	2.90	3.50	3.1410	0.153	0.0234	0.0244	4.8679	4.9709	95.90
Hemicellulose	23.61	26.12	24.916	0.668	0.2245	0.3445	1.9017	2.3557	65.17
Cellulose	27.73	30.83	29.361	0.889	0.8006	0.8314	3.0475	3.1055	96.30
DDM	62.50	64.97	63.581	0.695	0.4896	0.5076	1.1005	1.1206	96.45
DMI (% BW)	2.05	2.139	2.090	0.025	0.00058	0.0062	1.1554	1.1959	93.34
RFV	99.97	106.54	103.033	2.177	5.1778	5.1873	2.2084	2.2105	99.82

). Crude protein (CP), ash, crude fiber (CF), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), cellulose, digestible dry matter (DDM), dry matter intake (DMI), and relative feed value (RFV) generally exhibited high heritability, reflecting a predominant genetic influence. Hemicellulose showed moderate heritability, suggesting a greater environmental contribution.

These results suggest that selective breeding for improved nutritional and fiber quality traits in Dactylis glomerata populations is feasible. Populations with consistently high genetic values and stable performance across environments represent promising candidates for forage improvement and breeding programs.

3.4. Performance of Dactylis glomerata Populations Based on Duncan’s Multiple Range Test

Duncan’s multiple range test revealed significant differences among Dactylis glomerata populations for all evaluated forage quality traits (

Table 5. Mean comparison of Dactylis glomerata populations based on Duncan’s multiple range test for crude protein (CP), ash, crude fiber (CF), acid detergent fiber (ADF), neutral detergent fiber (NDF), acid detergent lignin (ADL), hemicellulose, cellulose, and digestible dry matter (DDM) (all expressed as %), dry matter intake (DMI, % of body weight), and relative feed value (RFV). This approach allowed identification of significant differences among populations and highlighted genotypes with superior forage quality potential. Values represent means across two experimental years (combined analysis).

Population	CP	Ash	CF	NDF	ADF	ADL	Hemicellulose	Cellulose	DDM	DMI (% bw)	RFV
Palaiochori	14.98 a	7.875 b	26.63 h	57.23 g	32.45 f	3.245 c	24.78 f	29.21 f	63.62 e	2.097 c	103.4 c
Gorgogyri	13.50 e	7.451 e	26.98 g	57.96 c	32.29 g	3.055 f	25.67 b	29.24 f	63.74 d	2.070 f	102.3 d
Xyloparoiko	13.04 f	7.267 f	27.45 e	56.31 i	31.51 i	2.951 h	24.80 f	28.56 g	64.36 b	2.131 a	106.3 a
Prinos	12.30 h	6.412 i	27.20 f	57.27 f	33.30 c	3.040 f	23.97 g	30.26 b	62.96 h	2.095 d	102.3 d
Pialeia	12.82 g	6.821 g	28.12 d	57.72 d	32.70 e	3.149 d	25.02 de	29.55 e	63.43 f	2.079 e	102.2 e
Prodromos	13.82 d	7.836 c	26.50 i	58.43 a	33.42 b	3.340 b	25.02 e	30.08 c	62.87 i	2.054 h	100.1 h
Filyra	14.08 c	8.267 a	28.42 c	56.7 h	31.65 h	3.440 a	25.05 d	28.21 h	64.25 c	2.116 b	105.4 b
Kefalopotamos	12.11 i	7.734 d	28.57 b	57.7 e	33.70 a	3.048 f	24.00 g	30.65 a	62.65 j	2.080 e	101.0 f
Kori	14.55 b	6.241 j	26.29 j	56.68 h	30.93 j	3.012 g	25.76 a	27.91 i	64.81 a	2.117 b	106.4 a
Kato Rachi	11.74 j	6.685 h	28.60 a	58.18 b	33.08 d	3.128 e	25.10 c	29.95 d	63.13 g	2.063 g	100.9 g

Different letters within each column indicate statistically significant differences among population means based on Duncan’s multiple range test at p ≤ 0.05.

). Populations such as Kori and Xyloparoiko generally exhibited higher digestible dry matter (DDM), dry matter intake (DMI), and relative feed value (RFV), indicating superior overall forage quality, even though Kori had slightly higher crude protein (CP) than Xyloparoiko, and Palaiochori actually showed the highest CP.

In contrast, populations like Prodromos and Kato Rachi tended to show lower values across several traits, reflecting comparatively lower nutritional potential. These differences among populations were consistent across most quality parameters, suggesting stable patterns of variation rather than isolated trait effects.

These results highlight trends in forage quality among populations and suggest which genotypes may be promising candidates for further evaluation in breeding programs aimed at improving high-quality and stable forage, particularly when multi-trait performance is considered rather than single-trait selection, without overgeneralizing suitability based solely on a single trait.

3.5. Correlation Analysis of Forage Quality Traits in Dactylis glomerata Populations

Correlation analysis among forage quality traits revealed clear patterns linking chemical composition to nutritional indices (Figure 2). Crude protein (CP) was positively associated with digestible dry matter (DDM), dry matter intake (DMI), and relative feed value (RFV), while fiber fractions (CF, NDF, ADF, cellulose) were negatively associated with these nutritional traits. RFV was strongly associated with DDM and DMI, indicating that populations with higher protein and lower fiber generally provide better forage quality.

These patterns highlight the relationships among protein and fiber components and their impact on forage quality. Observed correlations are consistent with expectations and can inform selection strategies; interpretation of genetic influence should be considered alongside heritability results.

3.6. Principal Component Analysis of Forage Quality Variation in Dactylis glomerata Populations

Principal Component Analysis (PCA) of forage quality traits explained 72.1% of the total variation, with PC1 accounting for 52.8% and PC2 for 19.3% (Figure 3). The score plot revealed a clear separation of populations into groups based on nutritional and fiber characteristics. Populations such as Xyloparoiko, Filyra, and Kori clustered on the positive side of PC1, associated with higher digestibility and feed value, whereas populations like Prodromos, Kefalopotamos, and Kato Rachi were positioned on the negative side, reflecting higher structural fiber content. Other populations occupied intermediate positions, indicating moderate forage quality.

The loading plot indicated that DMI, RFV, DDM, and CP were positively associated with PC1, while NDF, ADF, and cellulose were negatively associated. Thus, PC1 represents a gradient of forage quality, separating populations with higher nutritional value from those with higher fiber content. PC2 was mainly influenced by ADL and ash, highlighting additional variation in fiber composition and mineral content.

Overall, the PCA supports the identification of populations with superior forage quality, complementing the correlation and heritability analyses. Populations clustered on the positive side of PC1 can be considered promising candidates for selection in breeding programs aiming to improve digestibility and feed value in Dactylis glomerata.

3.7. Network Ordination of Forage Quality Relationships Among Dactylis glomerata Populations

The network ordination diagram (Figure 4) visualizes the similarity relationships among the ten Dactylis glomerata populations. Replicates of the same population formed distinct clusters, confirming the consistency of trait expression within populations. Populations such as Xyloparoiko and Kori grouped closely, reflecting high nutritional quality, whereas populations like Kato Rachi, Pialeia, and Kefalopotamos clustered together on the opposite side, associated with higher fiber content or intermediate forage quality.

This approach complements the correlation and PCA analyses by providing an intuitive visualization of overall population relationships, highlighting clusters of populations with similar forage profiles. Such patterns can inform the identification of superior or stable genotypes, as populations forming tight clusters with favorable traits may represent promising candidates for breeding programs.

3.8. Hierarchical Clustering of Forage Quality Traits in Dactylis glomerata Populations

Hierarchical clustering of forage quality traits (Figure 5) revealed two main clusters of variables: one including protein, hemicellulose, DDM, DMI, RFV, Ash, and ADL, and another consisting mainly of structural fiber components (CF, NDF, ADF, Cellulose).

Populations clustered according to overall trait profiles, with Xyloparoiko, Kori, and Filyra showing higher nutritional quality and lower fiber, while Prinos, Kefalopotamos, and Kato Rachi were associated with higher fiber content. The remaining populations displayed intermediate patterns.

These clusters complement the PCA and network ordination results by providing a clear grouping of populations based on combined trait performance. The identified patterns can help highlight populations with superior forage quality and stability for potential breeding selection.

4. Discussion

Considerable variation was observed among the ten Dactylis glomerata populations from the Trikala region in traits related to chemical composition, fiber fractions, and overall forage quality. Previous studies have documented substantial genotypic diversity in natural D. glomerata populations across Greece and other regions [15,42,43], providing a context for interpreting the observed differences. Such inherited variability provides a valuable foundation for future breeding programs, supporting the development of cultivars with enhanced nutritional value and stability under varying environmental conditions [44,45]. These patterns reflect the combined influence of genotypic potential and environmental responsiveness, indicating that populations differ in their nutritional profiles and in their response to environmental variability, highlighting relationships among traits and differentiation between populations.

Our combined ANOVA for Dactylis glomerata populations showed that genotypic effects were highly significant for all measured traits, including crude protein (CP), crude fiber (CF), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), digestible dry matter (DDM), and relative feed value (RFV). Environmental effects were also significant, although generally smaller than the genotypic effects, indicating that genotypic variability is a major determinant of forage quality. The genotype × environment interaction was significant across all traits, reflecting differential responses of populations under varying conditions. These findings support previous studies that emphasize the importance of genotypes and G × E interactions in maintaining stable forage quality in cocksfoot and other forage species [12,46,47,48,49].

High broad-sense heritability (H² > 93% for most traits, Table 4) confirms that these forage quality traits—particularly crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), and relative feed value (RFV)—are predominantly under genetic regulation and suggest strong potential for genetic improvement through selection. Hemicellulose, with moderate heritability (H² = 65.2%), appears more influenced by environmental conditions, highlighting the importance of multi-environment evaluation [26,50]. These findings align with previous studies reporting high heritability for CP, NDF, ADF, and RFV in cocksfoot and other forage species [12,47], and traits with higher H² observed in our study tended to show reduced environmental sensitivity and stronger inherited control compared with more plastic traits, indicating greater stability and predictability for selection [51,52,53]. Key traits, such as dry matter intake (DMI) and fiber fractions including NDF, are well-recognized parameters for evaluating forage quality and their potential impact on animal nutrition [25,54]. The relatively high heritability values observed here may also reflect the controlled experimental conditions and reduced environmental variability, as broad-sense heritability is defined as the ratio of genetic to total phenotypic variance, with lower environmental variation leading to higher estimates [55]. Overall, these results demonstrate that the measured forage quality traits in the studied Dactylis glomerata populations are under strong genotypic influence, while selection strategies should carefully consider environmental variability to optimize breeding efficiency. Heritability estimates should always be interpreted within the context of the experimental conditions under which they were obtained.

Populations Kori and Xyloparoiko consistently displayed higher values for key measured forage quality traits, including crude protein (CP), digestible dry matter (DDM), dry matter intake (DMI), and relative feed value (RFV), whereas Palaiochori was primarily characterized by its high CP content rather than overall performance across all traits. This distinction highlights that high protein content alone does not necessarily indicate superior forage quality, which is better captured by composite indices such as RFV. In contrast, populations such as Prodromos and Kato Rachi exhibited lower values for these nutritional traits. Stability Index analysis identified Kefalopotamos and Filyra as the most environmentally stable populations for DDM, DMI, and RFV, indicating consistent trait performance across diverse conditions. The use of SI in the present study provides a practical approach for preliminary evaluation of genotypes under variable environments. Although statistically significant differences among populations were detected, the relatively small numerical differences observed for several traits suggest limited biological differentiation under the experimental conditions. This may be partly attributed to the uniform crop management practices and relatively consistent environmental conditions across plots.

The observed variation in both quality and stability among populations is consistent with established knowledge on the inheritance of forage traits. Traits with high heritability, such as CP and RFV, tend to exhibit greater stability across environments, suggesting relatively simpler genetic control compared to more complex traits [48,56,57,58,59]. By contrast, more complex quantitative traits or those sensitive to environmental fluctuations, such as hemicellulose, exhibit lower Stability Index values and require evaluation across multiple environments for accurate selection [59,60,61,62]. Therefore, identifying populations that combine high-quality traits with environmental stability is critical for developing reliable cultivars suited to variable Mediterranean conditions.

Correlation analyses revealed strong associations among chemical composition, fiber fractions, and nutritional value. In particular, crude protein (CP) was positively correlated with digestible dry matter (DDM; r = 0.58 **) and relative feed value (RFV; r = 0.51 **), and negatively correlated with neutral detergent fiber (NDF; r = −0.36 **), acid detergent fiber (ADF; r = −0.58 **), and crude fiber (CF; r = −0.66 **). Similarly, RFV was strongly associated with DDM (r = 0.91 **) and dry matter intake (DMI; r = 0.92 **), indicating that populations with higher protein and lower fiber content exhibit improved nutritional indices. These patterns are consistent with previous studies on cocksfoot, where CP was generally negatively associated with fiber components, and DDM showed positive associations with quality traits [8,13,49,63]. Furthermore, research on ethnobotanically important grasses in Pakistan confirmed that protein content positively correlates with in vitro digestibility and negatively with fiber fractions, reinforcing the relevance of these traits for assessing forage nutritional quality [64]. The observed negative correlation between crude protein and fiber components (NDF and ADF) aligns with previous reports [26,65]. These relationships reflect the structural composition of the plant cell wall: NDF contributes to bulkiness, limiting dry matter intake due to rumen fill, while ADF is negatively related to digestibility. Acid detergent lignin (ADL) further restricts microbial degradation of cellulose and hemicellulose, reducing digestible dry matter (DDM) [65]. Together, these mechanisms explain the strong associations observed between fiber fractions and forage quality indices.

The observed trait correlations support targeted selection strategies: increasing CP while reducing NDF and ADF can simultaneously enhance digestibility and feed value. Populations exhibiting this combination, such as Palaiochori and Kori, represent valuable genetic material for future quality-focused selection and germplasm evaluation, rather than direct agronomic recommendation.

Multivariate analyses further supported the relationships observed in the correlation analysis. Principal component analysis, network ordination, and hierarchical clustering each contributed complementary insights into the relationships among traits, collectively highlighting a separation between forage quality traits and structural fiber fractions across the studied Dactylis glomerata populations. Traits associated with higher nutritional value, including crude protein (CP), digestible dry matter (DDM), dry matter intake (DMI), and relative feed value (RFV), were grouped together and positioned opposite to fiber-related traits such as neutral detergent fiber (NDF), acid detergent fiber (ADF), and cellulose. This pattern reflects the well-established trade-off between forage digestibility and structural fiber accumulation in grasses, where increased cell-wall components generally reduce intake potential and nutritional value. The limiting role of cell-wall concentration and lignification on forage digestibility and intake has been widely documented in ruminant nutrition studies [26,65]. Furthermore, variation in forage quality traits among populations may reflect both environmental influences and underlying genotypic effects, which are known to affect the nutritional characteristics of forage species [66]. Similar patterns of population differentiation and genetic variability have been reported in orchardgrass germplasm studies, highlighting the importance of natural populations as valuable genetic resources for forage improvement [20,67]. Moreover, the consistent grouping of populations Xyloparoiko, Filyra, and Kori across the PCA, network ordination, and hierarchical clustering analyses suggests a stable pattern of differentiation associated with higher forage quality traits. The convergence of these independent multivariate approaches strengthens the reliability of the observed relationships among traits and populations.

The high genotypic variance and heritability across traits suggest strong potential for selection and genetic improvement. Populations with superior nutritional quality and stability, such as Kefalopotamos, Filyra, Palaiochori, and Xyloparoiko, may serve as potential parental material in preliminary breeding or selection programs focused on forage quality improvement, rather than immediate cultivar development. Moreover, knowledge of population-specific trait stability allows the recommendation of particular genotypes for further experimental evaluation under multi-environment trials.

The study also demonstrates the utility of combined evaluation of both chemical composition and fiber fractions. By integrating SI, heritability, and trait correlations, breeders can select genotypes that optimize both nutritional quality and environmental resilience, improving forage availability and livestock performance.

From a breeding perspective, the combination of high crude protein and low fiber content (particularly NDF) is desirable, as it improves both digestibility and intake. However, selection based solely on nutritional quality may lead to trade-offs with yield and persistence, suggesting that balanced selection strategies are required.

Although dry matter yield is an important parameter in forage evaluation, the present study focuses specifically on forage quality traits in order to assess the genetic variability and phenotypic differentiation in nutritional traits among autochthonous Dactylis glomerata populations. In forage production systems, nutritional quality plays a critical role, particularly in ruminant feeding, where parameters such as crude protein, digestibility, and intake directly influence animal performance [26,54]. Moreover, forage quality and biomass yield are often not positively correlated, and may involve trade-offs, as negative or weak genetic correlations between dry matter yield and quality traits have been reported in forage grasses [13,51]. This highlights the importance of evaluating these components separately. Therefore, the results of this study provide valuable information on the variability of forage quality traits, while further research integrating yield and persistence would allow a more comprehensive agronomic evaluation.

In Mediterranean grazing systems, where environmental variability is high, the selection of genotypes that combine acceptable forage quality with stability across environments is particularly important. Populations identified as both high-performing and stable in this study may be suitable for use in low-input or stress-prone environments as genetic resources for future selection efforts.

In conclusion, Dactylis glomerata populations from the Trikala region exhibit considerable genetic variation for key forage quality traits. Populations such as Palaiochori, Xyloparoiko, Kori, Kefalopotamos, and Filyra demonstrated relatively high nutritional value and trait stability, indicating potential interest for further evaluation in germplasm conservation and quality-oriented breeding research. These findings provide a basis for assessing local germplasm and highlight the importance of conserving native populations as a resource for sustainable livestock production and future cultivar development, while acknowledging that additional studies on yield and field performance are needed to fully support practical applications.

5. Conclusions

Analysis of ten native Dactylis glomerata populations from the Trikala region revealed marked differences in chemical composition, fiber fractions, and overall forage quality. Populations such as Kori and Xyloparoiko exhibited higher values in several quality traits, while Filyra and Kefalopotamos were distinguished by higher environmental stability. Palaiochori was mainly characterized by high crude protein content rather than consistently superior performance across all traits.

High broad-sense heritability for key traits—including crude protein, fiber fractions, digestible dry matter, and relative feed value—indicates a strong genetic component underlying forage quality variation. Correlation analyses highlighted consistent relationships between protein content, fiber composition, and digestibility. Multivariate analyses further confirmed the differentiation among populations and the separation between nutritional and structural traits.

These results highlight substantial genetic variability among native populations and support their value as germplasm resources for future studies focused on forage quality traits and breeding programs targeting improved performance and adaptability.

Further evaluation of these populations—including yield, persistence, and multi-environment field performance—is required for comprehensive agronomic evaluation under diverse conditions and for identifying superior genotypes for practical forage production systems.