Exploring Phenological and Agronomic Parameters of Greek Lentil Landraces for Developing Climate-Resilient Cultivars Adapted to Mediterranean Conditions

Abstract

Lentil (Lens culinaris Medik. subsp. culinaris) is a Mediterranean legume crop of high value due to nutritional quality and adaptability; however, its cultivation is increasingly threatened due to climate uncertainty and reduction in genetic diversity in modern cultivars. The present research study evaluated 31 Greek lentil accessions (twenty-two landraces and nine commercial cultivars of both small and large seed types) in a semi-arid environment of Central Greece, over two cropping seasons, focusing on phenological, morphological, yield, and quality traits. The great diversity observed at the morpho-phenological and qualitative levels implies the high genotypic diversity of these genetic resources. Small-seeded landraces performed better in seed and biological yield, harvest index, and protein content, having greater phenological stability and tolerance to the Mediterranean environments. In particular, the highest seed yield was observed in LAX small-seeded landrace (1930 kg ha⁻¹), followed by TSO (1559 kg ha⁻¹), DIG (1449 kg ha⁻¹), and EGL (1437 kg ha⁻¹) small-seeded landraces. As for the regression analysis, seed yield was positively correlated with days to flowering (TF: r = 0.076, p < 0.01), plant height (PH: r = 0.143, p < 0.05), number of pods per plant (NPP: r = 0.941, p < 0.001), number of seeds per pod (NPP: r = 0.432, p < 0.001), number of branches (NPB: r = 0.234, p < 0.01), biological yield (BY: r = 0.683, p < 0.001), and harvest index (HI: r = 0.650, p < 0.001). Principal component analysis (PCA) distinguished small-seeded landraces associated with adaptive and yield traits from large-seeded cultivars associated with seed size. Greek lentil landraces, especially the small-seeded genotypes (e.g., LAX and DIG), have great potential for use in the development of climate-tolerant and high-yielding lentil varieties adapted for sustainable Mediterranean production. Breeding programs can target the crossing of landraces with large-seeded cultivars (e.g., IKAm and THEm) to develop varieties that combine stress tolerance, adaptation, and high productivity with adaptation to different seed sizes. Subsequent studies on drought tolerance and heat resistance are still important for continued improvement in lentil productivity in a changing climate.

1. Introduction

Lentil (Lens culinaris Medik. subsp. culinaris) is a self-pollinating and cool-season diploid (2n = 2x = 14) annual plant, and is said to be one of the prominent legumes for human and animal nutrition [1,2,3] since its seeds have recently been assigned the feature of functional food, owing to its high nutritional value content, polyphenols, and other bioactive compounds [4,5]. It is an older pulse, domesticated about 9000 years ago in the Near East’s “Fertile Crescent” and later diffused throughout the Mediterranean basin and Central Asia [1,6,7]. Nowadays, lentil is the fourth major pulse crop in the world after common bean, pea, and chickpea [6,8]. It is grown on an area of around 5.6 million hectares, with global production being approximately 7.07 million tonnes annually [9].

Of the lentil genus, Lens, there are eight taxa in four distinct species: L. nigricans, L. ervoides, L. lamottei, and L. culinaris with four subspecies (subsp. culinaris, subsp. orientalis, subsp. odemensis and subsp. tomentosus) [9]. L. culinaris subsp. culinaris, the sole cultivated taxon, is also divided into two type categories based on seed diameter and 100-seed weight: (a) the macrosperma type with diameter 6–9 mm and weight more than 4 g; and (b) the microsperma type with diameter 2–4 mm and weight less than 4 g [10,11].

Lentil is well-recognized for its high-quality protein, which ranges from 20% to 35% and includes a large number of amino acids, including lysine and isoleucine; hence, it is a cheap source of protein in many regions of the world both for humans as well as for animals [12,13]. Within the same species, the literature reports varying ranges of protein values that are ascribed to genetic variability and environmental factors, such as the year and region of cultivation [14,15,16]. It also provides slow-digesting carbohydrates, dietary fiber, low fat, a diversity of micronutrients, and bioactive health-promoting compounds, including polyphenols [12,17,18]. Due to its high iron content and low glycemic index, lentil is recommended for individuals with diabetes, obesity, and heart maladies [11,17].

Lentil cultivation can have a crucial position in developing Mediterranean farming systems because of its adaptability, mainly in marginal areas typified by rainfall scarcity and high temperatures [11]. By introducing lentils into crop rotations, the soil’s nitrogen and water availability are improved for the subsequent crop, potential soil erosion from wind and water is reduced, and soil properties are enhanced, with a higher level of nitrogen and organic carbon in the soil profile [19,20,21]. While these are positive, recent climatic change, especially in the Mediterranean region, has affected lentil yields, and this trend is to persist [22]. In Greece, domestic lentil production has sharply declined over the last three decades and is not sufficient to meet demand for use. Therefore, imports must be made due to decreased production caused by several factors, including lower profitability for farmers, crop diseases, and most importantly, unfavorable weather conditions that have resulted in drought [23]. Lentil cultivation in Greece increased to 11,500 ha in 2020, with the total production reaching 13,538 tons; however, the mean seed yield declined to less than 1.17 t ha⁻¹ [9,23]. Given the increasing likelihood of severe droughts and rising temperatures, it is critical to investigate techniques for adapting lentil production to these unpredictable circumstances [24], and the development of climate-resilient lentil cultivars constitutes one of these.

One of the major challenges in lentil improvement is the narrow genetic basis of the most common cultivars, which limits their ability to adapt and increase yield [25]. This situation is also related to the strict quality standards (purity, size and uniformity, color and appearance, moisture content, and defect limits on broken, split, or diseased lentils) established by both consumers and processors. To extend the genetic base while still maximizing selection advantages, it is critical to collect beneficial genes from different germplasm resources such as wild relatives and landraces [1,3,25].

Lentil landraces are a valuable genetic resource for diversity, specialized adaptation to particular environmental conditions, and genes that provide resistance to biotic and abiotic stress [3,26,27]. These landraces are a combination of diverse genotypes that evolved through natural and artificial selection. As a consequence, there is concern regarding lentil landraces as a source of genetic diversity [7,11]. Furthermore, landraces have been cultivated for decades in adverse environmental circumstances and low-input farming systems due to their substantial adaptability [1].

Morpho-phenological and agronomic characteristics are critical for identifying lentil genetic resources. Additionally, phenotypic parameters have been effectively employed for investigating genetic variation [23,26]. As a consequence, this characterization is critical and valuable for plant breeders seeking to improve lentil germplasm [24]. Consistently introducing new germplasm as a source of various genes is crucial for improving lentil cultivars [24,28]. Although parents with a greater genetic distance are capable of creating a hybrid with higher yield potential, current commercial cultivars are typically genetically similar and consist of a weak genetic base [29,30]. As a result, the phenotypic and genotypic diversity of lentil landraces and cultivars is required to establish their breeding efficiency potential for boosting crop production [11,29].

Recognizing the significance of landraces and the scarcity of knowledge about them, especially on lentil germplasm from Greece, this study was performed to explore variability in Greek lentil landraces collected from different regions of Greece by investigating phenological and agronomic characteristics for two consecutive years. The identification of novel and diverse genotypes from this material will contribute to the development of improved cultivars suited to future agricultural and climatic challenges under Mediterranean conditions.

2. Materials and Methods

2.1. Plant Material

The present research study evaluated 31 Greek lentil accessions acquired from various sources, including the following: six landraces from the USDA-ARS Western Regional Plant Introduction Station (Pullman, WA, USA); eight genotypes, specifically five landraces and three commercial cultivars from the Leibniz Institute of Plant Genetic and Crop Plant Research (IPK Gatersleben; Gatersleben, Germany); seven landraces from the ELGO-Dimitra Institute of Plant Breeding and Genetic Resources (Thermi, Thessaloniki, Greece); six commercial cultivars from the ELGO-Dimitra Institute of Industrial and Forage Crops (Larissa, Greece); and four landraces (unnamed accessions) which the Laboratory of Agronomy of the Agricultural University of Athens (Athens, Greece) collected directly from farmer fields. The nine commercial cultivars were used as testers in the evaluation of the agro-morphological traits.

More details on the plant material used in this experiment are presented in

Table 1. List of the thirty-one examined Greek lentil genotypes by genotype class, seed size class, accession number, accession name, date of collection, collecting institute, region, and collecting site.

Genotype			Accession Number	Landrace/Cultivar Name	Date of Collection	Collecting Institute	Region	Collecting Site
ID	Genotype Class	Seed Size
DOM_Ldr	Landrace	Large Seed	-	-	2017	Laboratory of Agronomy, Agricultural University of Athens	Central Greece	Domokos, Fthiotis
LA2 large	Landrace	Large Seed	PI 297749	ILL 281	12 May 1964	Western Regional Plant Introduction Station, USDA-ARS	Thessaly	Larissa
APK_Ldr	Landrace	Small Seed	-	-	2017	Laboratory of Agronomy, Agricultural University of Athens	Peloponnese	Arkadia
BEL	Landrace	Small Seed	K-095/06	-	22 September 2006	Institute of Plant Breeding and Genetic Resources, ELGO-Dimitra	Western Macedonia	Velanidia, Kozani
CHR	Landrace	Small Seed	LENS 65	LENS 65	1942	Leibniz Institute of Plant Genetic and Crop Plant Research	Peloponnese	Chrysovitsi, Arkadia
DIG	Landrace	Small Seed	IK-100/06	-	25 September 2006	Institute of Plant Breeding and Genetic Resources, ELGO-Dimitra	Ionian Islands	Digaleto, Kefallinia
DIL	Landrace	Small Seed	GRC1766/04	-	8 August 2004	Institute of Plant Breeding and Genetic Resources, ELGO-Dimitra	Thessaly	Dilofo, Larissa
EGL	Landrace	Small Seed	PI 633933	G075	9 September 1999	Western Regional Plant Introduction Station, USDA-ARS	Ionian Islands	Englouvi, Lefkada
EGL_Ldr	Landrace	Small Seed	-	-	2019	Laboratory of Agronomy, Agricultural University of Athens	Ionian Islands	Englouvi, Lefkada
GER	Landrace	Small Seed	HL-164/07	-	8 July 2007	Institute of Plant Breeding and Genetic Resources, ELGO-Dimitra	Crete	Tzermiado, Lasithi
GYT	Landrace	Small Seed	LENS 47	LENS 47	1942	Leibniz Institute of Plant Genetic and Crop Plant Research	Peloponnese	Polovitsa, Lakonia
IER	Landrace	Small Seed	LENS 83	LENS 83	1942	Leibniz Institute of Plant Genetic and Crop Plant Research	Crete	Ierapetra, Lasithi
KOR	Landrace	Small Seed	PI 297765	ILL 297	12 May 1964	Western Regional Plant Introduction Station, USDA-ARS	Western Macedonia	Korynos, Kastoria
KOZ	Landrace	Small Seed	LENS 7	LENS 7	1941	Leibniz Institute of Plant Genetic and Crop Plant Research	Western Macedonia	Monastiraki, Kozani
LA1	Landrace	Small Seed	PI 297739	ILL 271	12 May 1964	Western Regional Plant Introduction Station, USDA-ARS	Thessaly	Larissa
LAX	Landrace	Small Seed	F-173/06	-	4 October 2006	Institute of Plant Breeding and Genetic Resources, ELGO-Dimitra	Western Macedonia	Lachanokipoi, Kastoria
MOL	Landrace	Small Seed	PI 297790	ILL 322	12 May 1964	Western Regional Plant Introduction Station, USDA-ARS	Ionian Islands	Molle, Kefallinia
MOL_Ldr	Landrace	Small Seed	-	-	2017	Laboratory of Agronomy, Agricultural University of Athens	Ionian Islands	Molle, Kefallinia
PRG	Landrace	Small Seed	PI 297789	ILL 321	12 May 1964	Western Regional Plant Introduction Station, USDA-ARS	Ionian Islands	Pyrgi, Kefallinia
PTO	Landrace	Small Seed	LENS 4	LENS 4	1941	Leibniz Institute of Plant Genetic and Crop Plant Research	Western Macedonia	Agricultural Experimental Station of Ptolemaida, Kozani
RIZ	Landrace	Small Seed	P-118/06	-	25 August 2006	Institute of Plant Breeding and Genetic Resources, ELGO-Dimitra	Peloponnese	Riza, Lakonia
TSO	Landrace	Small Seed	K-106/06	-	23 September 2006	Institute of Plant Breeding and Genetic Resources, ELGO-Dimitra	Western Macedonia	Tsotyli, Kastoria
ELPm	Cultivar	Large Seed	-	ELPIDA	2019	Institute of Industrial and Forage Crops, ELGO-Dimitra (commercial cultivar—registration date: 19 September 2016)	-	-
IKAm	Cultivar	Large Seed	-	IKARIA	2019	Institute of Industrial and Forage Crops, ELGO-Dimitra (commercial cultivar—registration date: 27 July 1990)	-	-
THEm	Cultivar	Large Seed	-	THESSALIA	2019	Institute of Industrial and Forage Crops, ELGO-Dimitra (commercial cultivar—registration date: 7 March 1986)	-	-
ARK	Cultivar	Small Seed	LENS 235	ARCADIA	1989	Leibniz Institute of Plant Genetic and Crop Plant Research (Donor: Institute of Industrial and Forage Crops, ELGO-Dimitra) (commercial cultivar—registration date: 7 March 1986)	-	-
ARX	Cultivar	Small Seed	LENS 234	ARACHOVA	1989	Leibniz Institute of Plant Genetic and Crop Plant Research (Donor: Institute of Industrial and Forage Crops, ELGO-Dimitra) (commercial cultivar—registration date: 7 March 1986)	-	-
ATHm	Cultivar	Small Seed	-	ATHENA	2019	Institute of Industrial and Forage Crops, ELGO-Dimitra (commercial cultivar—registration date: 29 December 2006)	-	-
DIMm	Cultivar	Small Seed	-	DIMITRA	2019	Institute of Industrial and Forage Crops, ELGO-Dimitra (commercial cultivar—registration date: 7 March 1986)	-	-
PEL	Cultivar	Small Seed	LENS 237	PELASGIA	1989	Leibniz Institute of Plant Genetic and Crop Plant Research (Donor: Institute of Industrial and Forage Crops, ELGO-Dimitra) (commercial cultivar—registration date: 7 March 1986)	-	-
SAMm	Cultivar	Small Seed	-	SAMOS	2019	Institute of Industrial and Forage Crops, ELGO-Dimitra (commercial cultivar—registration date: 27 July 1990)	-	-

by genotype class (landrace or commercial cultivar), seed size (large or small seed, i.e., macrosperma or microsperma type, respectively), accession number, landrace or cultivar name, date of collection, collecting institute, as well as collection site and region of Greece. The qualitative characteristics of examined lentil genotypes, including presence of stem anthocyanin coloration, leaflet size, leaflet shape, standard color of flower, presence of violet stripes on the flower standard, main seed color and pattern of seed secondary color, are presented in

Table 2. Qualitative characteristics of the thirty-one examined lentil genotypes according to UPOV descriptors [31].

Genotype			Stem: Anthocyanin Coloration	Leaflet: Size	Leaflet: Shape	Flower: Color of Standard	Flower: Violet Stripes of Standard	Seed: Main Color	Seed: Pattern of Secondary Color *
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	Present	Large	Elliptic	Pink-Blue	Present	Green-Pink	Absent
LA2 large	L	LS	Absent	Large	Elliptic	White	Present	Green-Pink	Absent
APK_Ldr	L	SS	Present	Medium	Elliptic	Pink-Blue	Present	Green-Pink	Absent + Blotched
BEL	L	SS	Present	Small	Elliptic	Pink-Blue	Present	Green-Pink	Absent
CHR	L	SS	Present	Medium	Elliptic	White	Present	Pink-Green	Absent
DIG	L	SS	Absent	Medium	Elliptic	Blue	Present	Pink-Green	Absent + Marbled
DIL	L	SS	Absent	Large	Elliptic	White	Present	Green	Absent
EGL	L	SS	Absent	Medium	Elliptic	White	Present	Green-Pink	Absent
EGL_Ldr	L	SS	Absent	Medium	Elliptic	White	Present	Green-Pink	Absent + Blotched + Marbled-Blotched
GER	L	SS	Present	Medium	Elliptic	White	Present	Green	Absent
GYT	L	SS	Absent	Large	Elliptic	Pink-Blue	Present	Green	Absent
IER	L	SS	Absent	Medium	Obovate	White	Present	Green	Absent + Blotched + Marbled-Blotched
KOR	L	SS	Absent	Small	Elliptic	Pink-Blue	Present	Green	Absent
KOZ	L	SS	Present	Small	Elliptic	Pink-Blue	Present	Green	Absent
LA1	L	SS	Absent	Medium	Elliptic	White	Absent	Greenish Yellow	Absent
LAX	L	SS	Absent	Small	Elliptic	Pink-Blue	Present	Pink-Green	Absent + Blotched
MOL	L	SS	Present	Medium	Elliptic	Pink-Blue	Present	Pink-Green	Absent + Blotched
MOL_Ldr	L	SS	Absent	Medium	Elliptic	Blue-Pink	Present	Pink-Green	Absent + Blotched
PRG	L	SS	Absent	Medium	Elliptic	White	Present	Pink-Green	Absent
PTO	L	SS	Absent	Medium	Obovate	White	Present	Green	Absent
RIZ	L	SS	Absent	Small	Obovate	Pink-Blue	Present	Green	Absent + Blotched + Spotted + Marbled
TSO	L	SS	Present	Medium	Obovate	Pink-Blue	Present	Pink-Green	Absent
ELPm	C	LS	Absent	Medium–Large	Elliptic	White	Present	Green	Absent
IKAm	C	LS	Absent	Large	Elliptic	White	Present	Green	Absent
THEm	C	LS	Absent	Medium–Large	Elliptic	White	Present	Pink-Green	Absent
ARK	C	SS	Absent	Medium	Elliptic	White	Present	Green	Absent
ARX	C	SS	Absent	Medium	Elliptic	Pink-Blue	Present	Pink-Green	Absent
ATHm	C	SS	Absent	Medium	Elliptic	Pink-Blue	Present	Green	Absent
DIMm	C	SS	Absent	Medium	Elliptic	White	Present	Pink-Green	Absent
PEL	C	SS	Absent	Medium	Elliptic	White	Absent	Green-Pink	Absent
SAMm	C	SS	Present	Medium	Elliptic	White	Present	Green	Absent

* The extra characterizations (more than one separated by plus (+)) were recorded when the specific characteristics were found in more than 20% of the evaluated sample (sample size: 200 seeds).

according to the lentil descriptors of the International Union for the Protection of New Varieties of Plants (UPOV) [31]. In addition, images of the harvested seeds of the examined genotypes are also shown in Figure 1.

2.2. Study Location, Experimental Design, and Crop Management

The experiment was carried out at the experimental field of the Laboratory of Agronomy of the Agricultural University of Athens (AUA), located in Athens (37°59′ N, 23°42′ E, altitude: 29 m a.s.l.), Central Athens regional unit, Attica region, East–Central Greece for two consecutive years (2020/2021 and 2021/2022). The experimental site soil is classified as Cambisol [32]. The soil structure and chemical characteristics were based on a soil sample from five randomly selected points before crop cultivation at a depth of 0–30 cm. These were the following: clay loam (CL; 29.5% Clay, 34.9% Silt, and 35.6% Sand) with pH (1:1 H₂O) 7.48, total nitrogen 0.131%, available phosphorus-Olsen P 13.3 mg kg⁻¹ soil, available potassium 223 mg kg⁻¹ soil, 1.793% organic matter, and 14.78% CaCO₃. The average monthly air temperature and precipitation during the experimental periods were acquired from the automatic weather station (Davis Vantage Pro2 Weather Station; Davis Instruments Corporation, Hayward, CA, USA) located on the experimental field that hosted the trial and are shown in Figure 2. The mean temperature during the crop cycle (November–June) was 15.9 °C and 15.3 °C in 2020/2021 and 2021/2022 cultivation periods. February 2021 and January 2022 were the coldest months for each growing period, with the values of mean air monthly temperature being 11.3 °C and 8.6 °C, respectively. As for the total rainfall during the same period, it was 230 mm in 2020/2021 and 213 mm in 2021/2022 cultivation, respectively.

The experiment was laid out in a randomized complete block design (RCBD), with three replications. Each genotype was planted in a plot with four rows. The rows were 2 m long with 10 cm space between plants and 30 cm within rows. The soil was prepared by moldboard plowing at a 25 cm depth two days before sowing. In general, the trial was conducted under ordinary agricultural practices without using any herbicide, fungicide, pesticide, or fertilizer to study the actual genetic diversity. Seed sowing was performed on 13 and 1 November 2020 and 2021, respectively. Weeds were controlled by hand-hoeing before the canopy was closed. No pests or diseases affected the crop during the growing seasons. Harvesting was carried out between the last fortnight of May and the first fortnight of June at 75% maturity [11].

2.3. Sampling, Measurements, and Methods

According to the International Board for Plant Genetic Resources (IBPGR) and the International Center for Agricultural Research in the Dry Areas (ICARDA) descriptors [33] compared with different evaluated sources, the following parameters were recorded during the crop cycle: time to flowering (TF; days from sowing to when 50% of the lentil plants per plot are in flower), time to maturity (TM; days from sowing to when 90% of the pods per plot are golden brown), number of plants at harvest (NP), seed yield (SY), seed yield loss percentage (SYLP; based on the percentages of shriveled seeds), seed yield loss (SYL), biological yield (BY), harvest index (HI; HI = BY/SY), plant height (PH), number of branches per plant (NBP), leaf area per plant (LAP), number of pods per plant (NPP), number of seeds per pod (NSP), seed yield per plant (SYP), thousand seed weight (TSW), seed diameter (SD), and seed protein content (SPC). In particular, LAP was determined from three plants per plot at flowering time using an automatic leaf area meter (Delta-T Devices Ltd., Burwell, Cambridge, UK). TF, TM, NP, SY, SYLP, SYL, BY, and HI parameters were measured on the plants of two central rows of each plot. SYLP was estimated as the percentage of shriveled seeds from the two central rows of each plot. SYL was calculated by multiplying this percentage by the total SY. NBP, NPP, NSP, and SYP were determined on ten randomly selected plants from each plot. Finally, TSW and SPC were determined on an adequate seed sample received from each plot at harvest time. Specifically, TSW on one sample from each plot, whereas SPC was estimated by using the Kjeldahl method with the use of a Kjeltec 8400 autoanalyzer on three samples (~0.5 g) per plot (Foss Tecator AB, Höganas, Sweden) as described by [34].

2.4. Statistical Analysis

The quantitative data for each variable were tested for analysis of variance (ANOVA) to determine variability among genotypes, genotype classes, and seed size, and were reported over two years. Tukey’s honestly significant difference (HSD) test was used to separate the differences in means. To assess trait relationships and identify influential components among agro-morphological and quality parameters, Pearson’s correlation coefficient was used. Moreover, principal component analysis (PCA) examined the relationships among the evaluated parameters and enabled the visualization and weighting of each parameter in the characterization of the germplasm under study. In addition, the coefficients of variation (CV) were determined as the ratio of standard deviation to the mean expressed as percentage. JMP 18 statistical software (SAS Institute Inc., Cary, NC, USA) was used to conduct the above-mentioned analyses on the data gathered for phenological, morphological, yield-related, and quality parameters [11,34].

3. Results

3.1. Phenological Parameters

All the examined lentil genotypes significantly affected all the evaluated phenological parameters (

Table 3. Time to flowering (TF) and time to maturity (TM) as affected by year, lentil genotypes, genotype class, and seed size class.

			Time to Flowering (TF; Days After Sowing to Flowering)	Time to Maturity (TM; Days After Sowing to Maturity)
Year
2020/2021			146.9 ± 2.0 b	193.5 ± 0.5 b
2021/2022			159.7 ± 1.5 a	208.0 ± 0.6 a
Genotype
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	158.8 ± 2.6 bcdef	202.0 ± 1.8 cdefg
LA2 large	L	LS	156.8 ± 2.6 efg	201.8 ± 2.7 cdefg
APK_Ldr	L	SS	158.7 ± 3.0 bcdef	203.0 ± 3.6 bcde
BEL	L	SS	155.7 ± 3.7 fg	196.5 ± 4.7 h
CHR	L	SS	178.3 ± 0.9 a	211.7 ± 1.7 a
DIG	L	SS	155.2 ± 3.2 fgh	204.0 ± 4.0 bcd
DIL	L	SS	159.7 ± 1.1 bcdef	205.6 ± 3.6 bc
EGL	L	SS	159.3 ± 2.8 bcdef	202.7 ± 2.6 bcdef
EGL_Ldr	L	SS	158.3 ± 2.3 cdefg	202.7 ± 4.8 bcdef
GER	L	SS	156.2 ± 3.1 fg	198.5 ± 5.6 gh
GYT	L	SS	150.2 ± 3.4 hij	192.3 ± 2.2 ij
IER	L	SS	157.7 ± 2.8 efg	199.7 ± 3.2 efgh
KOR	L	SS	155.0 ± 3.6 fgh	201.7 ± 5.1 defg
KOZ	L	SS	159.8 ± 2.2 bcdef	201.7 ± 5.1 defg
LA1	L	SS	163.8 ± 2.0 b	204.5 ± 2.9 bcd
LAX	L	SS	153.2 ± 2.4 ghi	196.5 ± 4.7 h
MOL	L	SS	163.0 ± 1.5 bcd	202.0 ± 1.8 cdefg
MOL_Ldr	L	SS	155.8 ± 3.2 fg	197.5 ± 2.9 h
PRG	L	SS	155.0 ± 3.1 fgh	196.7 ±2.3 h
PTO	L	SS	157.0 ± 3.5 efg	202.3 ± 3.9 bcdef
RIZ	L	SS	156.2 ± 3.0 fg	199.0 ± 1.4 fgh
TSO	L	SS	157.8 ± 2.7 defg	202.0 ± 1.8 cdefg
ELPm	C	LS	94.2 ± 7.3 k	191.8 ± 2.3 j
IKAm	C	LS	156.5 ± 3.2 efg	205.5 ± 3.7 bcd
THEm	C	LS	154.8 ± 3.2 fgh	206.0 ± 3.6 b
ARK	C	SS	148.8 ± 4.7 ij	196.5 ± 4.7 h
ARX	C	SS	163.5 ± 1.6 bc	204.5 ± 2.9 bcd
ATHm	C	SS	147.2 ± 4.9 j	196.0 ± 4.5 hi
DIMm	C	SS	156.8 ± 3.1 efg	203.3 ± 4.4 bcde
PEL	C	SS	162.0 ± 1.8 bcde	204.0 ± 3.1 bcd
SAMm	C	SS	91.2 ± 5.5 k	191.8 ± 2.3 j
Source of Variance		df
FYear		1	1406.837 ***	3540.654 ***
FGenotype		30	314.398 ***	44.489 ***
FYear × Genotype		30	8.799 ***	17.771 ***
Genotype Class
L			158.1 ± 0.7 a	201.1 ± 0.8 a
C			141.5 ± 3.9 b	200.0 ± 1.3 a
Source of Variance		df
FYear		1	32.335 ***	282.214 ***
FGenotype Class		1	44.102 ***	2.038 ns
FYear × Genotype Class		1	1.642 ns	0.775 ns
Seed Size
LS			144.0 ± 5.1 b	201.5 ± 1.6 a
SS			155.0 ± 1.2 a	200.6 ± 0.7 a
Source of Variance		df
FYear		1	18.209 ***	152.71 ***
FSeed Size		1	11.057 **	0.405 ns
FYear × Seed Size		1	0.479 ns	1.325 ns

Note: All values are means ± standard errors (SE) for three replicates and two experimental years. L: landrace; C: cultivar; LS: large seed; SS: small seed; different letters within a column and among treatments of each evaluated factor indicate significant differences according to Tukey’s HSD test (p < 0.05); df: degrees of freedom; ^ns not significant (p > 0.05); **, *** significant at p < 0.01 and 0.001 levels of probability, respectively.

). The genotype that presented the greatest values of time to flowering (TF) and time to maturity (TM) was CHR, a small-seeded landrace, with two-year mean values being 178.3 and 211.7 days, respectively, whereas the lowest values of these parameters were observed at genotypes ELPm (TF and TM: 94.2 and 191.8 days, respectively) and SAMm (91.2 and 191.8 days), which are a large-seeded and a small-seeded commercial cultivar, respectively.

Moreover, the coefficients of variation (CV) for TF parameter varied from 1.1% to 19.0% among genotypes (

Table 4. Coefficients of variation (CV; %) for all evaluated parameters as affected by lentil genotypes, genotype class, and seed size class over two years.

			TF	TM	NP	SY	SYLP	SYL	BY	HI	PH	NBP	LAP	NPP	NSP	SYP	TSW	SD	SPC
Genotype
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	4.0	2.2	23.2	58.8	46.3	61.2	43.5	41.0	31.9	68.6	26.7	54.1	9.0	58.7	9.1	15.4	0.8
LA2 large	L	LS	4.1	3.3	36.8	56.8	97.5	45.2	46.2	45.5	24.5	38.2	23.7	50.6	18.6	56.9	5.2	5.7	1.3
APK_Ldr	L	SS	4.7	4.3	11.0	57.2	159.7	63.7	20.4	48.8	14.0	16.0	17.4	48.2	7.2	57.1	7.9	30.6	1.5
BEL	L	SS	5.9	5.9	10.1	57.2	57.6	60.6	30.2	31.7	13.4	24.3	29.8	51.8	8.6	57.3	11.9	3.4	3.2
CHR	L	SS	1.1	2.0	25.0	85.2	78.4	26.8	27.2	76.9	25.2	38.9	24.4	55.1	41.1	85.0	9.2	27.1	1.0
DIG	L	SS	5.0	4.8	17.4	58.7	63.1	90.5	69.2	15.3	19.1	39.5	21.5	64.9	3.6	58.6	11.2	3.2	2.4
DIL	L	SS	1.7	4.0	20.0	74.9	79.8	120.9	58.9	37.4	12.9	24.9	36.2	63.7	7.3	74.8	11.5	2.8	1.4
EGL	L	SS	4.3	3.1	12.1	41.7	49.7	39.4	23.1	26.2	19.0	23.8	23.1	39.3	8.4	41.6	6.3	4.5	0.6
EGL_Ldr	L	SS	3.5	5.8	13.3	36.1	29.0	30.4	18.9	27.4	21.9	21.8	15.6	36.1	4.3	36.2	6.7	4.8	3.5
GER	L	SS	4.9	6.9	12.5	62.8	89.1	57.1	25.9	68.4	17.2	20.7	23.6	51.5	9.7	62.7	11.0	3.0	4.0
GYT	L	SS	5.6	2.8	22.7	52.7	59.0	74.4	48.6	58.7	9.6	9.4	27.2	52.8	8.8	52.8	13.0	5.5	1.6
IER	L	SS	4.3	3.9	19.9	56.6	49.6	59.4	23.6	79.2	11.6	23.2	16.0	58.7	9.2	56.5	22.2	8.7	1.7
KOR	L	SS	5.7	6.2	11.4	60.2	75.1	91.0	32.3	58.0	25.7	24.3	15.0	54.4	13.9	60.3	11.7	16.8	3.8
KOZ	L	SS	3.3	6.2	20.7	31.8	78.0	84.6	36.8	16.2	23.7	33.3	21.6	29.5	4.4	31.6	7.2	2.5	3.6
LA1	L	SS	3.0	3.5	41.0	79.1	62.3	99.6	35.1	83.8	16.1	52.5	22.7	67.2	21.3	79.2	13.9	8.2	0.9
LAX	L	SS	3.8	5.9	12.2	15.2	37.9	47.7	15.5	16.9	16.4	9.6	20.0	11.2	5.7	15.3	7.1	3.5	2.8
MOL	L	SS	2.3	2.2	17.0	57.0	41.2	65.2	43.5	36.5	20.0	35.0	13.0	54.1	7.9	57.1	16.5	4.1	0.7
MOL_Ldr	L	SS	5.0	3.6	40.9	46.6	73.4	94.4	36.5	22.3	10.9	42.4	19.6	50.8	4.7	46.7	12.1	3.8	0.9
PRG	L	SS	5.0	2.8	35.2	80.6	71.3	117.4	53.7	33.9	6.1	41.2	24.7	77.1	10.2	80.8	12.6	1.9	2.1
PTO	L	SS	5.5	4.7	17.7	47.7	48.9	47.9	22.9	43.9	14.6	16.8	16.7	38.2	7.7	47.6	12.2	5.9	2.3
RIZ	L	SS	4.7	1.7	24.8	31.0	32.7	40.2	34.5	13.9	24.9	16.2	17.8	34.3	10.3	31.1	12.0	4.5	2.8
TSO	L	SS	4.1	2.2	14.1	34.3	53.4	63.7	24.5	19.2	20.0	19.3	21.7	32.3	3.1	34.4	6.6	3.3	0.9
ELPm	C	LS	19.0	3.0	60.5	114.6	103.4	64.1	61.0	68.0	14.8	28.9	20.0	114.8	15.9	114.5	18.1	5.7	1.6
IKAm	C	LS	4.6	4.0	69.1	62.8	96.5	96.1	47.0	28.2	13.1	30.4	6.5	57.3	8.5	62.7	11.0	8.6	1.6
THEm	C	LS	5.1	4.3	13.3	44.5	39.3	72.9	43.4	34.4	21.3	26.5	32.0	41.9	4.3	44.6	13.6	4.0	1.6
ARK	C	SS	7.8	5.9	8.9	54.2	43.1	64.5	13.3	54.8	16.5	11.9	17.3	51.5	10.3	54.3	16.3	3.3	3.2
ARX	C	SS	2.5	3.5	21.0	87.2	54.7	119.1	44.3	66.2	28.6	41.3	27.6	83.9	10.6	87.4	10.0	9.7	0.8
ATHm	C	SS	8.1	5.6	21.9	47.1	55.5	56.7	41.3	25.4	19.2	13.0	16.2	44.2	6.2	47.2	13.8	14.6	3.0
DIMm	C	SS	5.0	5.3	21.4	44.9	121.0	100.7	31.4	37.6	29.7	26.1	10.9	44.0	17.7	44.8	10.5	3.9	2.6
PEL	C	SS	2.7	3.8	9.1	101.4	59.8	70.9	19.2	89.6	22.8	36.3	11.1	86.3	13.8	101.3	33.8	5.4	1.2
SAMm	C	SS	14.8	3.0	28.0	36.4	77.4	46.4	21.1	28.5	5.3	60.3	20.0	35.2	5.2	36.3	9.8	5.5	0.8
Genotype Class
L			5.0	4.4	24.5	65.1	117.6	80.1	41.4	49.5	22.2	38.3	25.5	8.8	17.3	65.0	33.8	17.7	2.9
C			20.4	4.9	30.2	65.7	92.3	89.0	52.4	56.1	20.3	41.5	29.5	13.9	11.4	65.9	34.5	16.2	3.2
Seed Size
LS			19.2	4.2	42.9	72.1	93.5	84.8	64.5	50.6	20.9	66.5	32.4	67.4	12.2	72.3	12.2	8.9	2.6
SS			10.1	4.6	22.0	67.8	112.9	80.2	40.9	51.6	21.5	33.8	25.0	65.9	16.6	67.6	23.6	12.1	3.1

Note: L: landrace; C: cultivar; LS: large seed; SS: small seed. Studied parameters: TF: time to flowering; TM: time to maturity; NP: number of plants at harvest; SY: seed yield; SYLP: seed yield loss percentage; SYL: seed yield loss; BY: biological yield; HI: harvest index; PH: plant height; NBP: number of branches per plant; LAP: leaf area per plant; NPP: number of pods per plant; NSP: number of seeds per pod; SYP: seed yield per plant; TSW: thousand seed weight; SD: seed diameter; SPC: seed protein content.

), indicating a very low to medium degree of variability, while the corresponding CV for TM varied from 1.7% to 6.9%, presenting a very low to low variability.

At the genotype class level, it was observed that only TF was significantly influenced, with the mean values being 158.1 and 141.5 days in landraces and cultivars, respectively (Table 3). Moreover, for landraces, the CV was 5.0%, which can be considered a high degree of stability, whereas the corresponding CV for cultivars was 20.4%. At the seed size level, the large-seeded genotypes presented a significantly lower value of TF (144.0 days) as compared to small-seeded genotypes (155.0 days). On the contrary, there were no substantial variances across landraces and cultivars, and large-seeded and small-seeded genotypes on the TM parameter. The growing season had a considerable influence on phenological parameters, with the TF and TM presenting the lowest values in 2020/2021 (146.9 and 193.5 days) as compared to the subsequent year (2021/2022: 159.7 and 208.0 days, respectively), which might be attributed to difference in weather conditions, with the first year presenting higher mean monthly air temperatures during vegetative stage (December–March) in comparison to the second experimental year (Figure 2).

3.2. Plant Population and Yield Performance

The number of plants at harvest differed substantially among genotypes, with BEL and LAX small-seeded landraces achieving the highest two-year mean values (30.0 and 29.6 plants m⁻²), and the LA2 large, a large-seeded landrace, presented the lowest value (14.1 plants m⁻²). In the same manner, there were also significant differences across genotype classes and seed size types, with the greatest values observed in landraces (23.7 plants m⁻²) and small-seeded genotypes (24.0 plants m⁻²).

Concerning seed yield, significant differences were determined among all evaluated genotypes, with the lowest value for LA1 small-seeded landrace (272 kg ha⁻¹), the highest for LAX small-seeded landrace (1930 kg ha⁻¹), followed by TSO (1559 kg ha⁻¹), DIG (1449 kg ha⁻¹), and EGL (1437 kg ha⁻¹) small-seeded landraces (

Table 5. Number of plants at harvest (NP), seed yield (SY), seed yield loss percentage (SYLP), and seed yield loss (SYL) as affected by year, lentil genotypes, genotype class, and seed size class.

			Number of Plants at Harvest (NP; Plants per m2)	Seed Yield (SY; kg ha−1)	Seed Yield Loss Percentage (SYLP; %)	Seed Yield Loss (SYL; kg ha−1)
Year
2020/2021			22.7 ± 0.7 a	899 ± 65 a	8.6 ± 1.0 a	47 ± 3 a
2021/2022			23.6 ± 0.6 a	826 ± 58 a	6.8 ± 0.7 a	45 ± 4 a
Genotype
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	20.7 ± 1.9 abcde	1231 ± 296 bcdef	8.7 ± 1.6 bc	102 ± 26 a
LA2 large	L	LS	14.1 ± 2.1 e	765 ± 177 fghijklm	14.6 ± 5.8 abc	67 ± 12 abc
APK_Ldr	L	SS	27.8 ± 1.2 ab	858 ± 200 defghijk	7.4 ± 4.8 bc	33 ± 9 bc
BEL	L	SS	30.0 ± 1.2 a	1252 ± 293 bcde	2.4 ± 0.6 c	28 ± 7 bc
CHR	L	SS	25.6 ± 2.6 abc	291 ± 101 mn	23.4 ± 7.5 a	44 ± 5 abc
DIG	L	SS	25.2 ± 1.8 abc	1437 ± 344 bc	2.8 ± 0.7 c	49 ± 18 abc
DIL	L	SS	23.0 ± 1.9 abcde	328 ± 100 lmn	11.9 ± 3.9 abc	52 ± 25 abc
EGL	L	SS	25. 2 ± 1.2 abc	1449 ± 247 bc	4.7 ± 1.0 c	62 ± 10 abc
EGL_Ldr	L	SS	25.2 ± 1.4 abcd	1324 ± 195 bcd	6.3 ± 0.8 c	80 ± 9 abc
GER	L	SS	26.3 ± 1.3 abc	547 ± 140 hijklmn	9.7 ± 3.5 abc	33 ± 8 bc
GYT	L	SS	22.6 ± 2.1 abcde	788 ± 170 efghijkl	4.9 ± 1.2 bc	40 ± 12 abc
IER	L	SS	22.2 ± 1.8 abcde	536 ± 124 hijklmn	18.8 ± 3.8 ab	95 ± 23 ab
KOR	L	SS	23.7 ± 1.1 abcde	627 ± 154 hijklmn	11.6 ± 3.6 abc	56 ± 21 abc
KOZ	L	SS	21.9 ± 1.8 abcde	1265 ± 164 bcde	2.5 ± 0.8 c	34 ± 12 bc
LA1	L	SS	17.8 ± 3.0 cde	272 ± 88 n	8.5 ± 2.2 bc	19 ± 7 c
LAX	L	SS	29.6 ± 1.5 a	1930 ± 120 a	1.9 ± 0.3 c	39 ± 8 abc
MOL	L	SS	23.0 ± 1.6 abcde	945 ± 220 defghi	3.5 ± 0.6 c	33 ± 9 bc
MOL_Ldr	L	SS	22.2 ± 3.7 abcde	884 ± 168 defghij	2.5 ± 0.7 c	21 ± 8 c
PRG	L	SS	25.9 ± 1.9 abc	1013 ± 334 cdefgh	1.7 ± 0.5 c	17 ± 8 c
PTO	L	SS	21.1 ± 2.1 abcde	754 ± 147 fghijklm	7.5 ± 1.5 bc	50 ± 10 abc
RIZ	L	SS	27.8 ± 1.6 ab	1109 ± 140 bcdefg	3.2 ± 0.4 c	36 ± 6 abc
TSO	L	SS	15.6 ± 3.8 de	1559 ± 218 ab	3.1 ± 0.7 c	53 ± 14 abc
ELPm	C	LS	16.8 ± 5.9 cde	387 ± 181 klmn	12.9 ± 5.4 abc	20 ± 5 c
IKAm	C	LS	23.0 ± 1.2 abcde	533 ± 159 hijklmn	3.2 ± 1.6 c	20 ± 10 c
THEm	C	LS	25.9 ± 0.9 abc	707 ± 128 ghijklmn	9.4 ± 1.5 bc	71 ± 21 abc
ARK	C	SS	20.8 ± 1.8 abcde	906 ± 201 defghij	6.7 ± 1.2 bc	56 ± 15 abc
ARX	C	SS	22.2 ± 1.9 abcde	462 ± 164 jklmn	5.7 ± 1.3 c	30 ± 15 bc
ATHm	C	SS	24.1 ± 2.1 abcd	901 ± 173 defghij	5.5 ± 1.3 bc	51 ± 12 abc
DIMm	C	SS	24.1 ± 0.9 abcd	650 ± 119 ghijklmn	11.9 ± 5.9 abc	63 ± 26 abc
PEL	C	SS	19.3 ± 2.2 bcde	488 ± 202 ijklmn	13.6 ± 3.3 abc	41 ± 12 abc
SAMm	C	SS	20.7 ± 1.9 abcde	454 ± 68 jklmn	6.7 ± 2.1 bc	24 ± 5 c
Source of Variance		df
FYear		1	2.523 ns	1.184 ns	3.257 ns	0.092 ns
FGenotype		30	4.431 ***	5.827 ***	4.139 ***	2.994 ***
FYear × Genotype		30	3.782 ***	2.182 **	2.636 ***	2.316 ***
Genotype Class
L			23.7 ± 0.5 a	23.7 ± 0.5 a	7.3 ± 0.8 a	47 ± 3 a
C			21.6 ± 0.9 b	21.6 ± 0.9 b	8.5 ± 1.1 a	43 ± 5 a
Source of Variance		df
FYear		1	3.992 ns	0.074 ns	3.292 ns	0.125 ns
FGenotype Class		1	5.292 *	14.049 ***	1.022 ns	0.319 ns
FYear × Genotype Class		1	6.872 **	1.284 ns	1.990 ns	0.181 ns
Seed Size
LS			18.5 ± 1.4 b	18.5 ± 1.4 b	10.1 ± 1.8 a	58 ± 9 a
SS			24.0 ± 0.4 a	24.0 ± 0.4 a	7.2 ± 0.7 a	44 ± 3 b
Source of Variance		df
FYear		1	3.560 ns	1.391 ns	2.663 ns	0.029 ns
FSeed Size		1	23.491 ***	1.480 ns	2.945 ns	3.978 *
FYear × Seed Size		1	2.712 ns	5.342 *	5.338 *	0.168 ns

Note: All values are means ± standard errors (SE) for three replicates and two experimental years. L: landrace; C: cultivar; LS: large seed; SS: small seed; different letters within a column and among treatments of each evaluated factor indicate significant differences according to Tukey’s HSD test (p < 0.05); df: degrees of freedom; ^ns not significant (p > 0.05); *, **, *** significant at p < 0.05, 0.01, and 0.001 levels of probability, respectively.

). In addition, TSO achieved the lowest CV (34.3%) (Table 4). The interaction of year × genotype was significant at p < 0.001 (Table 5). In addition, at the genotype classes level, the landraces presented significantly higher seed yield (962 kg ha⁻¹) compared to cultivars (614 kg ha⁻¹).

The seed yield loss percentage varied significantly only among the different genotypes (Table 5). Specifically, most genotypes showed significant variations, with CHR small-seeded landrace having the greatest percentage (23.4%). As for the seed yield loss, the highest yield (102 kg ha⁻¹) was recorded in DOM_Ldr large-seeded landrace, while the lowest values were obtained from SAMm (24 kg ha⁻¹) small-seeded cultivar, ELPm (20 kg ha⁻¹) and IKAm (20 kg ha⁻¹) large-seeded cultivars, as well as MOL_Ldr (21 kg ha⁻¹), LA1 (19 kg ha⁻¹), and PRG (17 kg ha⁻¹) small-seeded landraces. Furthermore, substantial differences were also observed between large- and small-seeded genotypes with the greatest value of seed yield loss (58 kg ha⁻¹) recorded in large-seeded genotypes, presenting an increase of 31.82% compared to the small-seeded genotypes.

As shown in

Table 6. Biological yield (BY) and harvest index (HI) as affected by year, lentil genotypes, genotype class, and seed size class.

			Biological Yield (BY; kg ha−1)	Harvest Index (HI)
Year
2020/2021			4759 ± 221 a	0.187 ± 0.010 a
2021/2022			4442 ±207 a	0.190 ± 0.011 a
Genotype
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	6979 ± 1239 ab	0.167 ± 0.028 efghi
LA2 large	L	LS	4782 ± 901 cdefghi	0.155 ± 0.029 fghij
APK_Ldr	L	SS	4778 ± 398 cdefghi	0.172 ± 0.034 efghi
BEL	L	SS	6438 ± 794 abc	0.190 ± 0.025 defgh
CHR	L	SS	3970 ± 441 hijkl	0.075 ± 0.024 j
DIG	L	SS	4521 ± 1277 efghij	0.346 ± 0.022 a
DIL	L	SS	3032 ± 729 jklm	0.104 ± 0.016 ij
EGL	L	SS	7212 ± 680 a	0.195 ± 0.021 defgh
EGL_Ldr	L	SS	5999 ± 463 abcde	0.216 ± 0.024 cdefg
GER	L	SS	4444 ± 470 efghij	0.132 ± 0.037 hij
GYT	L	SS	4021 ± 797 ghijkl	0.218 ± 0.052 cdefg
IER	L	SS	4268 ± 412 fghijk	0.147 ± 0.047 ghij
KOR	L	SS	5217 ± 687 cdefghi	0.126 ± 0.030 hij
KOZ	L	SS	4623 ± 695 efghij	0.284 ± 0.019 abc
LA1	L	SS	2593 ± 371 klm	0.107 ± 0.037 ij
LAX	L	SS	5997 ± 379 abcde	0.326 ± 0.022 ab
MOL	L	SS	4220 ± 749 fghijk	0.215 ± 0.032 cdefg
MOL_Ldr	L	SS	3613 ± 538 ijkl	0.232 ± 0.021 cdef
PRG	L	SS	4695 ± 368 defghij	0.096 ± 0.035 ij
PTO	L	SS	3594 ± 788 ijkl	0.256 ± 0.034 bcd
RIZ	L	SS	5817 ± 543 abcdef	0.133 ± 0.024 hij
TSO	L	SS	4193 ± 590 fghijk	0.267 ± 0.015 abcd
ELPm	C	LS	6393 ± 638 abcd	0.238 ± 0.019 cde
IKAm	C	LS	1703 ± 424 m	0.222 ± 0.062 cdefg
THEm	C	LS	2350 ± 521 lm	0.206 ± 0.026 cdefgh
ARK	C	SS	5689 ± 1007 abcdefg	0.130 ± 0.018 hij
ARX	C	SS	5662 ± 306 abcdefgh	0.159 ± 0.036 efghi
ATHm	C	SS	4036 ± 730 ghijkl	0.105 ± 0.028 ij
DIMm	C	SS	5274 ± 889 bcdefghi	0.171 ± 0.018 efghi
PEL	C	SS	4132 ± 529 fghijkl	0.159 ± 0.024 efghi
SAMm	C	SS	4695 ± 368 defghij	0.096 ± 0.035 ij
Source of Variance		df
FYear		1	1.898 ns	0.093 ns
FGenotype		30	5.306 ***	5.783 ***
FYear × Genotype		30	2.152 **	1.769 *
Genotype Class
L			12.8 ± 0.4 a	4850 ± 175 a
C			9.9 ± 0.6 b	3942 ± 284 b
Source of Variance		df
FYear		1	0.137 ns	0.325 ns
FGenotype Class		1	7.720 **	2.994 ns
FYear × Genotype Class		1	1.259 ns	0.376 ns
Seed Size
LS			12.0 ± 1.4 a	4389 ± 526 a
SS			11.9 ± 0.3 a	4627 ± 151 a
Source of Variance		df
FYear		1	0.091 ns	3.248 ns
FSeed Size		1	0.204 ns	0.644 ns
FYear × Seed Size		1	0.633 ns	5.438 *

Note: All values are means ± standard errors (SE) for three replicates and two experimental years. L: landrace; C: cultivar; LS: large seed; SS: small seed; different letters within a column and among treatments of each evaluated factor indicate significant differences according to Tukey’s HSD test (p < 0.05); df: degrees of freedom; ^ns not significant (p > 0.05); *, **, *** significant at p < 0.05, 0.01, and 0.001 levels of probability, respectively.

, biological yield was significantly affected by different studied genotypes, with the greatest values acquired from EGL small-seeded landrace (7212 kg ha⁻¹), whereas the lowest yields were obtained from ELPm (1703 kg ha⁻¹) large-seeded cultivar and SAMm (1567 kg ha⁻¹) small-seeded cultivar. Moreover, there were substantial differences among landraces and cultivars, with the landraces having a mean biological yield of 4850 kg ha⁻¹, presenting an increase of 23.03% compared to the commercial cultivars.

Regarding harvest index (HI), the results of the current study showed that there were significant differences among all evaluated genotypes, with the values ranging from 0.075 to 0.346, with the lowest value observed in CHR small-seeded landrace and the highest in DIG small-seeded landrace (Table 6).

In addition, according to Table 5 and Table 6, the growing season did not present significant differences between the two studied growing periods in number of plants at harvest (NP), seed yield (SY), biological yield (BY), harvest index (HI), seed yield loss (SYL), and its percentage (SYLP).

3.3. Yield-Related Parameters

As for the plant height (PH), considerable differences were determined among all studied genotypes (

Table 7. Plant height (PH), number of branches per plant (NBP), and leaf area per plant (LAP) as affected by year, lentil genotypes, genotype class, and seed size class.

			Plant Height (PH; cm)	Number of Branches per Plant (NBP)	Leaf Area per Plant (LAP; cm2)
Year
2020/2021			21.3 ± 0.4 b	10.8 ± 0.5 b	649 ± 19 a
2021/2022			27.0 ± 0.5 a	13.1 ± 0.4 a	574 ± 14 b
Genotype
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	26.1 ± 3.4 abc	19.3 ± 5.4 a	747 ± 82 abcd
LA2 large	L	LS	25.0 ± 2.5 abcde	13.9 ± 2.2 bcd	728 ± 70 abcde
APK_Ldr	L	SS	25.7 ± 1.5 abc	11.3 ± 0.7 cde	663 ± 47 cdefgh
BEL	L	SS	26.3 ± 1.4 abc	13.6 ± 1.4 bcd	533 ± 65 hijkl
CHR	L	SS	18.2 ± 1.9 e	12.9 ± 2.1 bcde	551 ± 55 ghijkl
DIG	L	SS	23.7 ± 1.8 abcde	10.0 ± 1.6 de	610 ± 53 defghijk
DIL	L	SS	22.2 ± 1.2 abcde	8.5 ± 0.9 ef	583 ± 86 fghijk
EGL	L	SS	22.4 ± 1.7 abcde	14.0 ± 1.4 bcd	724 ± 68 abcdef
EGL_Ldr	L	SS	25.7 ± 2.3 abc	13.2 ± 1.2 bcd	586 ± 37 efghijk
GER	L	SS	29.2 ± 2.0 a	13.4 ± 1.1 bcd	517 ± 50 ijkl
GYT	L	SS	25.5 ± 1.0 abcd	11.1 ± 0.4 cde	588 ± 65 efghijk
IER	L	SS	22.0 ± 1.0 abcde	10.9 ± 1.0 cde	835 ± 54 ab
KOR	L	SS	25.4 ± 2.7 abcde	14.4 ± 1.4 bcd	469 ± 29 kl
KOZ	L	SS	26.3 ± 2.5 abc	11.3 ± 1.5 cde	608 ± 54 defghijk
LA1	L	SS	20.4 ± 1.3 bcde	11.0 ± 2.4 cde	661 ± 61 cdefghi
LAX	L	SS	26.6 ± 1.8 abc	10.6 ± 0.4 cde	529 ± 43 hijkl
MOL	L	SS	19.5 ± 1.6 cde	16.6 ± 2.4 ab	656 ± 35 cdefghi
MOL_Ldr	L	SS	20.2 ± 0.9 bcde	14.7 ± 2.5 bc	573 ± 46 ghijk
PRG	L	SS	19.8 ± 0.5 cde	14.4 ± 2.4 bcd	591 ± 59 efghijk
PTO	L	SS	28.5 ± 1.7 a	10.5 ± 0.7 cde	784 ± 53 abc
RIZ	L	SS	18.4 ± 1.9 de	13.3 ± 0.9 bcd	541 ± 39 ghijkl
TSO	L	SS	25.1 ± 2.0 abcde	12.1 ± 1.0 bcde	752 ± 67 abcd
ELPm	C	LS	27.1 ± 1.6 ab	5.0 ± 0.6 f	404 ± 33 l
IKAm	C	LS	26.8 ± 1.5 abc	10.7 ± 1.4 cde	876 ± 29 a
THEm	C	LS	27.7 ± 2.4 a	11.1 ± 1.2 cde	690 ± 102 bcdefg
ARK	C	SS	25.7 ± 1.7 abc	10.5 ± 0.5 cde	562 ± 40 ghijk
ARX	C	SS	20.4 ± 2.4 bcde	12.2 ± 2.0 bcde	620 ± 70 defghij
ATHm	C	SS	24.0 ± 1.9 abcde	12.3 ± 0.7 bcde	504 ± 33 jkl
DIMm	C	SS	24.1 ± 2.9 abcde	11.7 ± 1.2 cde	535 ± 24 hijkl
PEL	C	SS	26.4 ± 2.5 abc	12.6 ± 1.9 bcde	589 ± 27 efghijk
SAMm	C	SS	24.9 ± 0.5 abcde	4.3 ± 1.1 f	402 ± 36 l
Source of Variance		df
FYear		1	142.801 ***	14.785 ***	16.351 ***
FGenotype		30	5.183 ***	3.161 ***	4.418 ***
FYear × Genotype		30	2.116 **	1.394 ns	1.056 ns
Genotype Class
L			23.7 ± 0.4 a	12.8 ± 0.4 a	628 ± 14 a
C			25.2 ± 0.7 a	9.9 ± 0.6 b	567 ± 23 b
Source of Variance		df
FYear		1	53.962 ***	9.038 **	4.237 *
FGenotype Class		1	3.273 ns	15.757 ***	5.190 *
FYear × Genotype Class		1	0.371 ns	0.002 ns	3.426 ns
Seed Size
LS			26.6 ± 1.0 a	12.0 ± 1.4 a	679 ± 42 a
SS			23.7 ± 0.4 b	11.9 ± 0.3 a	600 ± 12 b
Source of Variance		df
FYear		1	35.704 ***	2.969 *	4.036 *
FSeed Size		1	8.372 **	0.003 ns	5.778 *
FYear × Seed Size		1	0.089 ns	0.849 ns	0.137 ns

Note: All values are means ± standard errors (SE) for three replicates and two experimental years. L: landrace; C: cultivar; LS: large seed; SS: small seed; different letters within a column and among treatments of each evaluated factor indicate significant differences according to Tukey’s HSD test (p < 0.05); df: degrees of freedom; ^ns not significant (p > 0.05); *, **, *** significant at p < 0.05, 0.01, and 0.001 levels of probability, respectively.

), with the highest values for GER (29.2 cm) and PTO (28.5 cm) small-seeded landraces and THEm (27.7 cm) large-seeded cultivar, while the lowest PH was observed in CHR small-seeded landrace (18.2 cm). Similarly, there were also substantial differences among seed size types, with the greatest PH observed in large-seeded genotypes (26.6 cm), presenting an increase of 12.24% compared to small-seed genotypes, whereas the growing season also had a significant effect on this studied parameter, with the greatest mean value (27.0 cm) recorded during the second cropping season.

The number of branches per plant (NBP) varied significantly across evaluated genotypes (Table 7), with the greatest value for DOM_Ldr large-seeded landrace (19.3) and the lowest for ELPm large-seeded cultivar (5.0) and SAMm large-seeded cultivar (4.3). Likewise, there were substantial differences among genotype classes, with the cultivars presenting a significantly lower mean value of NBP (9.9) as compared to landraces (12.8). In addition, the growing season had a considerable effect on NPB, with the greatest value (13.1) recorded in 2021/2022, presenting an increase of 21.30% compared to the previous cultivation season.

According to Table 7, the leaf area per plant (LAP) significantly differed between the genotypes, with the highest two-year mean value recorded in IKAm large-seeded commercial cultivar (876 cm²), whereas the lowest LAP values were obtained from ELPm (404 cm²) large-seeded and SAMm (402 cm²) small-seeded commercial cultivars. The analysis on variance (ANOVA) on genotype class and seed size levels presented that there were significant differences among cultivars and landraces, and small- and large-seeded genotypes, with the greatest values found in landraces (628 cm²) and large-seeded genotypes (679 cm²), respectively. Regarding the growing season, this also had a substantial influence on the LAP, with the mean value of the first cultivation period (2020/2021: 649 cm²) being higher than that of the second period (2021/2022: 574 cm²).

According to the present research study’s results, the number of pods per plant (NPP) and the number of seeds per pod (NSP) were also affected by different lentil genotypes (

Table 8. Number of pods per plant (NPP), number of seeds per pod (NSP), and seed yield per plant (SYP) as affected by year, lentil genotypes, genotype class, and seed size class.

			Number of Pods per Plant (NPP)	Number of Seeds per Pod (NSP)	Seed Yield per Plant (SYP; g)
Year
2020/2021			66.6 ± 4.9 a	1.42 ± 0.03 b	2.68 ± 0.19 a
2021/2022			55.9 ± 4.1 b	1.47 ± 0.02 a	2.47 ± 0.17 a
Genotype
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	52.0 ± 11.5 fghijk	1.28 ± 0.05 hijkl	3.69 ± 0.89 bcdef
LA2 large	L	LS	33.7 ± 7.0 hijkl	1.18 ± 0.09 kl	2.29 ± 0.53 fghijklm
APK_Ldr	L	SS	60.8 ± 12.0 efghi	1.54 ± 0.05 bcdefghi	2.58 ± 0.60 defghijk
BEL	L	SS	99.4 ± 21.0 bcd	1.54 ± 0.04 bcdefghi	3.76 ± 0.88 bcde
CHR	L	SS	37.8 ± 8.5 hijkl	1.24 ± 0.21 jkl	0.87 ± 0.30 mn
DIG	L	SS	112.0 ± 29.7 abc	1.83 ± 0.03 ab	4.31 ± 1.03 bc
DIL	L	SS	27.7 ± 8.6 ijkl	1.31 ± 0.04 ghijkl	1.15 ± 0.31 lmn
EGL	L	SS	110.3 ± 17.7 abc	1.51 ± 0.05 cdefghij	4.35 ± 0.74 bc
EGL_Ldr	L	SS	102.5 ± 15.1 bcd	1.49 ± 0.03 cdefghij	3.97 ± 0.59 bcd
GER	L	SS	40.6 ± 8.5 hijkl	1.44 ± 0.06 defghijk	1.64 ± 0.42 hijklmn
GYT	L	SS	52.9 ± 11.4 fghijk	1.34 ± 0.05 fghijkl	2.36 ± 0.51 efghijkl
IER	L	SS	45.1 ± 11.6 ghijkl	1.14 ± 0.04 l	1.58 ± 0.45 hijklmn
KOR	L	SS	42.0 ± 9.3 ghijkl	1.57 ± 0.09 bcdefgh	1.88 ± 0.46 hijklmn
KOZ	L	SS	80.2 ± 9.7 cdef	1.63 ± 0.03 abcde	3.79 ± 0.49 bcde
LA1	L	SS	21.1 ± 5.8 kl	1.39 ± 0.12 efghijkl	0.82 ± 0.26 n
LAX	L	SS	137.5 ± 6.3 a	1.91 ± 0.04 a	5.79 ± 0.36 a
MOL	L	SS	88.1 ± 19.5 bcde	1.70 ± 0.05 abcd	2.84 ± 0.66 defghi
MOL_Ldr	L	SS	73.6 ± 15.3 defg	1.62 ± 0.03 abcdef	2.65 ± 0.51 defghij
PRG	L	SS	81.7 ± 25.7 cdef	1.77 ± 0.07 abc	3.04 ± 0.99 cdefgh
PTO	L	SS	53.9 ± 8.4 fghij	1.44 ± 0.05 defghijk	2.26 ± 0.44 fghijklm
RIZ	L	SS	83.7 ± 11.7 bcdef	1.60 ± 0.07 bcdefg	3.33 ± 0.42 bcdefg
TSO	L	SS	115.0 ± 15.2 ab	1.42 ± 0.02 defghijkl	4.68 ± 0.65 ab
ELPm	C	LS	18.4 ± 8.6 l	1.32 ± 0.09 ghijkl	1.16 ± 0.54 klmn
IKAm	C	LS	22.5 ± 6.1 jkl	1.26 ± 0.05 ijkl	1.69 ± 0.48 hijklmn
THEm	C	LS	31.1 ± 5.3 ijkl	1.24 ± 0.02 jkl	2.12 ± 0.39 ghijklmn
ARK	C	SS	65.8 ± 13.8 efgh	1.29 ± 0.05 hijkl	2.72 ± 0.60 defghij
ARX	C	SS	30.3 ± 12.5 ijkl	1.38 ± 0.07 efghijkl	1.28 ± 0.59 jklmn
ATHm	C	SS	59.2 ± 10.7 efghi	1.32 ± 0.03 ghijkl	2.70 ± 0.52 defghij
DIMm	C	SS	55.9 ± 10.0 efghij	1.23 ± 0.09 jkl	1.95 ± 0.36 ghijklmn
PEL	C	SS	42.8 ± 15.1 ghijkl	1.34 ± 0.08 fghijkl	1.47 ± 0.61 ijklmn
SAMm	C	SS	20.8 ± 3.0 kl	1.45 ± 0.03 defghijk	1.36 ± 0.20 jklmn
Source of Variance		df
FYear		1	6.319 **	5.995 *	1.219 ns
FGenotype		30	7.537 ***	12.830 ***	5.634 ***
FYear × Genotype		30	2.447 ***	3.973 ***	2.154 **
Genotype Class
L			71.0 ± 4.0 a	1.49 ± 0.02 a	2.91 ± 0.15 a
C			39.4 ± 3.9 b	1.31 ± 0.01 b	1.84 ± 0.17 b
Source of Variance		df
FYear		1	4.828 **	5.133 *	0.029 ns
FGenotype Class		1	21.887 ***	22.542 ***	14.479 ***
FYear × Genotype Class		1	1.864 ns	0.682 ns	1.055 ns
Seed Size
LS			32.1 ± 4.0 b	1.26 ± 0.03 b	2.21 ± 0.29 a
SS			67.7 ± 3.6 a	1.48 ± 0.02 a	2.68 ± 0.14 a
Source of Variance		df
FYear		1	4.023 **	4.469 *	1.146 ns
FSeed Size		1	18.14 ***	22.695 ***	1.899 ns
FYear × Seed Size		1	1.034 ns	2.239 ns	5.197 *

Note: All values are means ± standard errors (SE) for three replicates and two experimental years. L: landrace; C: cultivar; LS: large seed; SS: small seed; different letters within a column and among treatments of each evaluated factor indicate significant differences according to Tukey’s HSD test (p < 0.05); df: degrees of freedom; ^ns not significant (p > 0.05); *, **, *** significant at p < 0.05, 0.01, and 0.001 levels of probability, respectively.

). Specifically, the highest NPP and NSP values were in LAX (137.5 and 1.91, respectively) small-seeded landrace. In the same manner, there were significant differences between genotype classes, with the landraces presenting a substantially higher two-year mean value of NPP and NSP (71.0 and 1.49, respectively) as compared to commercial cultivars (39.4 and 1.31, respectively). In addition, there were also significant differences between seed size classes, with the small-seeded genotypes presenting a substantially higher two-year mean value of NPP and NSP (67.7 and 1.48, respectively) as compared to commercial cultivars (32.1 and 1.26, respectively). The growing season also had a significant impact on the NPP and NSP values, with the first cultivation period having the greatest NPP value (2020/2021: 66.6), and the second cultivation period presenting the highest NSP value (2021/2022: 1.47).

The seed yield per plant (SYP) as influenced by different lentil genotypes is shown in Table 8. The genotype that presented the highest value of SYP was LAX, a small-seeded landrace, with the two-year mean value being 5.79 g, while the lowest value of SYP was recorded at genotype LA1 (0.82 g), a small-seeded landrace. Furthermore, there were substantial differences across genotype classes and seed size types, with the highest SYP values seen in landraces (2.91 g) and small-seeded genotypes (2.68 g), whereas the growing season had no significant impact on the SYP.

3.4. Seed Physical and Quality Parameters

Concerning the thousand seed weight (TSW), the research results confirmed that this trait was significantly affected from the evaluated genotypes, with the lowest value for CHR small-seeded landrace (18.44 g), and the highest for IKAm and THEm large-seeded cultivars (56.50 and 55.20 g, respectively), followed by DOM_Ldr (54.94 g) and LA2 large (53.77 g) large-seeded landraces (

Table 9. Thousand seed weight (TSW), seed diameter (SD), and seed protein content (SPC) as affected by year, lentil genotypes, genotype class, and seed size class.

			Thousand Seed Weight (TSW; g)	Seed Diameter (SD; mm)	Seed Protein Content (SPC; %)
Year
2020/2021			30.52 ± 1.21 a	4.94 ± 0.08 a	21.76 ± 0.07 b
2021/2022			31.79 ± 1.20 a	4.83 ± 0.10 a	22.23 ± 0.06 a
Genotype
ID	Genotype Class	Seed Size
DOM_Ldr	L	LS	54.94 ± 2.04 a	6.35 ± 0.40 ab	21.92 ± 0.07 bcdef
LA2 large	L	LS	53.77 ± 1.15 a	6.91 ± 0.16 a	22.06 ± 0.12 bcdef
APK_Ldr	L	SS	26.11 ± 0.84 defghi	3.86 ± 0.48 f	22.42 ± 0.13 b
BEL	L	SS	24.26 ± 1.18 efghi	4.39 ± 0.06 def	21.50 ± 0.28 efg
CHR	L	SS	18.44 ± 0.69 i	3.94 ± 0.44 ef	23.35 ± 0.10 a
DIG	L	SS	22.29 ± 1.02 fghi	4.26 ± 0.06 def	22.44 ± 0.22 b
DIL	L	SS	30.93 ± 1.45 cde	4.82 ± 0.06 cdef	22.47 ± 0.13 b
EGL	L	SS	25.86 ± 0.67 defghi	4.61 ± 0.09 def	22.14 ± 0.06 bcd
EGL_Ldr	L	SS	26.09 ± 0.71 defghi	4.67 ± 0.09 def	22.18 ± 0.32 bcd
GER	L	SS	26.27 ± 1.18 defghi	4.70 ± 0.06 cdef	21.91 ± 0.36 bcdef
GYT	L	SS	33.89 ± 1.80 cd	4.99 ± 0.11 cd	21.05 ± 0.14 g
IER	L	SS	31.60 ± 3.03 cde	4.95 ± 0.18 cd	21.76 ± 0.15 cdef
KOR	L	SS	27.14 ± 1.29 cdefgh	4.57 ± 0.31 def	22.15 ± 0.34 bcd
KOZ	L	SS	28.93 ± 0.85 cdefg	4.95 ± 0.05 cd	22.08 ± 0.32 bcde
LA1	L	SS	26.71 ± 1.52 defghi	4.70 ± 0.16 cdef	22.40 ± 0.08 b
LAX	L	SS	21.98 ± 0.64 ghi	4.23 ± 0.06 def	21.77 ± 0.25 cdef
MOL	L	SS	18.71 ± 1.26 hi	3.96 ± 0.07 ef	22.19 ± 0.07 bcd
MOL_Ldr	L	SS	22.83 ± 1.12 efghi	4.31 ± 0.07 def	21.48 ± 0.08 efg
PRG	L	SS	20.64 ± 1.06 ghi	4.11 ± 0.03 def	21.73 ± 0.18 def
PTO	L	SS	28.58 ± 1.42 cdefg	4.79 ± 0.12 cdef	22.23 ± 0.21 bcd
RIZ	L	SS	25.84 ± 1.27 defghi	4.55 ± 0.08 def	21.98 ± 0.25 bcdef
TSO	L	SS	28.41 ± 0.76 cdefg	4.63 ± 0.06 def	22.18 ± 0.09 bcd
ELPm	C	LS	48.32 ± 3.57 ab	6.59 ± 0.15 ab	21.03 ± 0.14 g
IKAm	C	LS	56.50 ± 2.72 a	6.31 ± 0.24 ab	22.35 ± 0.17 bc
THEm	C	LS	55.20 ± 3.05 a	6.37 ± 0.10 ab	22.36 ± 0.15 bc
ARK	C	SS	30.48 ± 2.02 cdef	4.95 ± 0.07 cd	21.51 ± 0.28 efg
ARX	C	SS	25.61 ± 1.04 defghi	4.40 ± 0.19 def	22.34 ± 0.07 bcd
ATHm	C	SS	35.24 ± 1.99 c	4.72 ± 0.28 cdef	21.45 ± 0.26 fg
DIMm	C	SS	28.15 ± 1.21 cdefg	4.91 ± 0.08 cde	22.27 ± 0.24 bcd
PEL	C	SS	23.04 ± 3.18 efghi	4.52 ± 0.10 def	22.34 ± 0.11 bc
SAMm	C	SS	45.12 ± 1.81 b	5.66 ± 0.13 bc	20.98 ± 0.07 g
Source of Variance		df
FYear		1	3.181 ns	3.435 ns	132.31 ***
FGenotype		30	51.274 ***	19.773 ***	19.551 ***
FYear × Genotype		30	1.897 **	1.248 ns	7.296 ***
Genotype Class
L			28.31 ± 0.83 b	4.69 ± 0.07 b	22.06 ± 0.06 a
C			38.20 ± 1.80 a	5.38 ± 0.12 a	21.84 ± 0.09 b
Source of Variance		df
FYear		1	0.045 ns	0.742 ns	25.234 ***
FGenotype Class		1	27.736 ***	20.567 ***	4.937 *
FYear × Genotype Class		1	0.370 ns	0.107 ns	0.832 ns
Seed Size
LS			53.49 ± 1.21 a	6.50 ± 0.11 a	21.94 ± 0.11 a
SS			26.98 ± 0.51 b	4.58 ± 0.04 b	22.01 ± 0.05 a
Source of Variance		df
FYear		1	0.009 ns	1.005 ns	7.112 **
FSeed Size		1	444.027 ***	287.328 ***	0.925 ns
FYear × Seed Size		1	0.533 ns	0.186 ns	2.198 ns

Note: All values are means ± standard errors (SE) for three replicates and two experimental years. L: landrace; C: cultivar; LS: large seed; SS: small seed; different letters within a column and among treatments of each evaluated factor indicate significant differences according to Tukey’s HSD test (p < 0.05); df: degrees of freedom; ^ns not significant (p > 0.05); *, **, *** significant at p < 0.05, 0.01, and 0.001 levels of probability, respectively.

). At the genotype class and seed size levels, it was found that TSW was significantly affected, with the highest two-year mean values being 38.20 and 53.49 g in cultivars and large-seeded genotypes, respectively.

Seed diameter (SD) was also substantially influenced by different lentil genotypes, with the greatest value for the LA2 large (6.91 mm), a large-seeded landrace, and the lowest for APK_Ldr (3.86 mm), a small-seeded landrace (Table 9). Moreover, the corresponding CV for SD of LA2 large and APK_Ldr genotypes was 5.7% and 30.6%, presenting low and moderate variability, respectively (Table 4). SD was also shown to be highly impacted at the genotype class and seed size levels, with the greatest two-year mean values in cultivars and large-seeded genotypes being 5.38 and 6.50 mm, respectively. Furthermore, as in TSW, SD was not affected by the growing season (p > 0.05; Table 8).

Finally, concerning seed protein content (SPC), the findings of the present research study confirmed that this parameter was influenced by genotypes (Table 9). In particular, the highest SPC was found in CHR (23.35%) small-seeded landrace, whereas the lowest values were recorded in ELPm (21.03%) large-seeded cultivar, and SAMm (20.98%) small-seeded cultivar. At the genotype class level, the landraces presented a significantly higher value of SPC (22.06%) as compared to cultivars (21.84%). The growing season also had a considerable effect on SPC, presenting a higher mean value in 2021/2022 (22.23%) as compared to the 2020/2021 growing period (21.76%).

3.5. Principal Component Analysis (PCA) of Evaluated Parameters

The results of principal component analysis (PCA), including factor loading matrix, eigenvalues, variability, and cumulative percentage for all examined parameters, are shown in

Table 10. Principal component analysis (PCA): factor loadings, eigenvalues, variability (%), and cumulative (%) of the studied lentil parameters.

Parameters	Factor-1	Factor-2	Factor-3	Factor-4	Factor-5
TF	0.216	0.775	−0.232	0.119	−0.291
TM	−0.057	0.741	−0.075	0.590	0.127
NP	0.375	0.288	−0.194	−0.377	0.439
SY	0.955	−0.125	0.189	0.033	0.021
SYLP	−0.474	0.416	0.284	−0.421	0.214
SYL	0.334	0.332	0.657	−0.201	0.150
BY	0.711	0.343	0.384	−0.249	0.081
HI	0.590	−0.549	−0.038	0.340	−0.026
PH	0.146	0.225	0.325	0.471	0.554
NBP	0.330	0.453	0.141	0.134	−0.446
LAP	0.149	0.247	0.413	−0.249	−0.504
NPP	0.952	−0.047	0.015	−0.118	−0.006
NSP	0.549	−0.261	−0.559	0.307	−0.027
SYP	0.945	−0.120	0.188	0.042	0.021
TSW	−0.313	−0.289	0.738	0.363	−0.034
SD	−0.279	−0.297	0.765	0.288	−0.094
SPC	−0.047	0.800	−0.117	0.329	−0.018
Eigenvalue	4.74	3.17	2.58	1.64	1.14
Variability (%)	27.89	18.62	15.20	9.63	6.72
Cumulative (%)	27.89	46.51	61.71	71.34	78.06

Note: Statistically high weights are bolded. TF: time to flowering; TM: time to maturity; NP: number of plants at harvest; SY: seed yield; SYLP: seed yield loss percentage; SYL: seed yield loss; BY: biological yield; HI: harvest index; PH: plant height; NBP: number of branches per plant; LAP: leaf area per plant; NPP: number of pods per plant; NSP: number of seeds per pod; SYP: seed yield per plant; TSW: thousand seed weight; SD: seed diameter; SPC: seed protein content.

. The PCA generated five main factors (eigenvalue > 1), which explained 78.06% of the total variance of the studied lentil traits. The factors accounted for 27.89%, 18.62%,15.20%, 9.63%, and 6.72% of the total variability, from the first to the fifth, respectively. The first factor has a strong positive loading (factor loadings > 0.70) for SY, BY, and NPP, a moderate positive loading (factor loadings between 0.40 and 0.70) for HI and NSP, and a moderate negative loading for SYLP. The second factor had a strong positive loading for TF and TM, a moderate positive for SYLP and NBP, and a moderate negative for HI. As for the third factor, a positive strong loading was found for the TSW and SD, a moderate positive for SYL and LAP, and a moderate negative for NSP. In the fourth factor, a moderate positive loading was observed for TM and PH, and a moderate negative loading for SYLP. Finally, in the fifth factor, a positive moderate loading was found for the PH and NP, and a moderate negative loading for NBP and LAP.

PCA biplot was used to extract relevant information from the examined dataset and highlight similarities and differences between genotype classes and seed size classes. The PCA biplot was conducted on the whole dataset after grouping the variables by genotype, genotype class, and seed size type as shown in Figure 3. The mean values were then determined for all evaluated genotypes over the two years and the three replications. Most of the variation was accounted for by PC1 (27.9%), which distinguished most small-seeded landraces (on the positive side) from most large-seeded cultivars (on the negative side), while the PC2 revealed 18.6% of the total variance and discriminated landraces and small-seeded genotypes (on the positive site) from the cultivars and large-seeded genotypes (on the negative site). Small-seeded landraces, including APK_Ldr, MOL, EGL, EGL_Ldr, TSO accessions, and the large-seeded landrace DOM_Ldr, were located in the upper-right quarter and were strongly associated with the variables TF, NP, LAP, PH, NBP, SYL, and BY, which contribute positively to both components. Small- and large-seeded cultivars, including IKAm, ATHm, ELPm and SAMm, and a small-seeded landrace (GYT) were located in the lower-left quarter and were associated with two variables (TSW and SD) contributing negatively to both components. Seven small-seeded landraces (BEL, DIG, KOZ, RIZ, MOL_Ldr, PRG, and LAX) and one small-seeded cultivar (ARK) were located in the third quadrant and were associated with SY, SYP, NSP, and HI parameters. Finally, the genotypes GER, THEm, KOR, PTO, DIMm, IER, ARX, PEL, DIL, LA1, LA2 large, and CHR were located in the upper-left quarter and were more associated with SPC, TM, and SYLP parameters.

4. Discussion

Local lentil landraces are of major importance in indigenous germplasm evaluation, especially under recent climate changes. Such genotypes are among the best genetic resources as they exhibit good tolerance to stressful environments [6,35,36]. Even though many lentil landraces have been placed in foreign gene banks, they have yet to be utilized fully in breeding [37,38]. Twenty-two Greek lentil landraces (twenty small-seeded and two large-seeded) from six districts (Western Macedonia, Thessaly, Central Greece, Ionian Islands, Peloponnese, and Crete), as well as nine Greek commercial varieties (three small-seeded and six large-seeded), were investigated for seventeen morphological, phenological, and agronomic characteristics in this research. Results indicated higher genetic diversity in landraces than in commercial cultivars, suggesting the occurrence of useful genetic diversity and desirable gene combinations that can be utilized for the improvement of lentil [3,39]. This diversity, as a result of environmental adaptation and natural or human-driven selection under various environments, emphasizes the potential of Greek landraces in further enhancing adaptability and resilience in future breeding [40,41].

The current research indicated a huge genetic variability between all genotypes tested for all the evaluated phenological characters, illustrating the significant impact of genetic components on the duration of the developmental phases in lentils [42,43]. The huge ranges observed (91.2–178.3 days and 191.8–211.7 days for time to flowering (TF) and time to maturity (TM), respectively) evidence the existence of early- and late-maturing types, as expression of adaptation to various agro-ecological environments [43,44,45]. Specifically, the smallest-seeded landrace CHR had the longest biological cycle, while the largest-seeded cultivar ELPm and the smallest-seeded cultivar SAMm had the shortest. This variation is consistent with the observation of Erskine et al. [45], who found that flowering and maturity time variation between lentil genotypes is dominated primarily by temperature response rather than other environmental factors like photoperiod, indicating that lentil phenology is particularly sensitive to thermal conditions. Lower TF and TM values in the 2020/2021 season compared to 2021/2022 were likely due to higher mean temperatures during the vegetative period, promoting crop development. A similar phenological response caused by temperature has been previously noted in lentil and other cool-season legumes [46,47].

The differences between landraces and cultivars, and between seed size types, that were captured during observation are indicative of varying adaptive strategies. Landraces and small-seeded types had a longer phenological cycle, presumably because they had adapted to low-input traditional production systems and cooler conditions, while commercial cultivars and large-seeded types, developed for high-input and favorable conditions, matured quicker to escape terminal drought [45,48]. It should also be noted that the large-seeded landraces do not strictly follow this pattern, since they are, in general, adapted to a wider range of environments rather than developed for intensive production systems [34]. Early flowering and short vegetative period are especially advantageous in Mediterranean crop conditions where terminal droughts are increasingly becoming common. This practice has been documented in several herbaceous crops, such as wheat and certain grain legumes [11,49]. Lentil yield loss can be up to 24% when the reproductive stage is stressed, and up to 70% when the pod-filling stage is stressed [11].

Early-maturing types are most essential under the Mediterranean climate, in which legumes are exposed to heat stress as well as drought during grain filling. Such stresses have worsened with increased frequency and severity due to the ongoing effects of climate change. Moroccan research has shown that heat, as well as drought, affected flowering and agronomic traits and the nutrition of lentils by shortening their growth periods [50]. Similar findings have also been observed in several legumes such as lentil, common bean, chickpea, and faba bean [11,50,51,52].

The large genotypic diversity of plant population density (NP) at harvest and seed yield (SY) is also an indicator of the genetic variability of the analyzed lentil germplasm. LAX and BEL landraces had the highest plant population densities, which are indicative of better seedling establishment and adaptability, as previously reported that landraces are more resistant to stress and present a reasonable yield under low-input farming systems [11,53]. Seed yield was quite variable (272–1930 kg ha⁻¹) and higher in small-seeded landraces compared to large-seeded cultivars. In addition, Parihar et al. [54] also reported similar trends, and that in the case of small-seeded landraces, they exhibit higher reproduction efficiency and stability compared to others in suboptimal conditions. Although the growing season did not exert a significant main effect, the high-order genotype × year interaction indicates that yield performance was influenced by specific environmental factors, particularly temperature and rainfall distribution [46].

Higher harvest index (HI) and biological yield (BY) demonstrated by landraces over cultivars signify their greater biomass productivity and assimilate partitioning efficiency. These results validate Ninou et al. [55], who emphasized the high-yielding potential and adaptive character sources of landraces. Higher loss of seed yield revealed by large-seeded genotypes confirms their higher pod shattering susceptibility and lower harvest efficiency, which was already documented by Erskine [56].

To successfully initiate a crop improvement program, it is important to realize the interrelationship between various traits. Selection for advanced traits such as yield, which is usually determined later in the development stage, is likely to lengthen the breeding cycle and slow genetic progress [57,58]. Since yield is a quantitative trait with both genetic and environmental control, indirect selection is a more feasible alternative [59]. Indirect selection for traits that are easier to measure and highly correlated with yield can promote selection efficiency [59,60].

SY in this work was positively correlated with days to flowering (TF: r = 0.076, p < 0.01), plant height (PH: r = 0.143, p < 0.05), number of pods per plant (NPP: r = 0.941, p < 0.001), number of branches (NPB: r = 0.234, p < 0.01), biological yield (BY: r = 0.683, p < 0.001), and harvest index (HI: r = 0.650, p < 0.001) (Figure 4). Yield in indeterminate legumes relies on traits like the number of branches, the number of pods, biomass, and the ratio of flowers that develop to pods. The positive correlated relationship between most of the traits indicates that multiple-trait improvement may be simultaneous. Attention towards such yield-contributing traits will therefore result in drastic seed yield improvement. The same was reported in previous research [61,62,63].

However, SY was poorly negatively correlated with thousand seed weight (TSW: r = −0.098, p < 0.05), and perhaps choosing for this character alone will lead to reduced efficiency or genetic gain. Furthermore, TSW was negatively correlated with seeds per pod (NSP: r = −0.407, p < 0.001), and NSP was positively correlated with SY (r = 0.432, p < 0.001). These same weak correlations between SY and TSW have been reported in other research studies [61,64], In contrast, studies by Kumar et al. [61] and Sharma et al. [65] indicate a positive correlation between these two parameters.

As lentil crops in the Mediterranean region are likely to face terminal heat and drought stress due to climate fluctuations, the production of short-duration cultivars is necessary to limit yield losses [64]. The significant correlation coefficient (r) values between TF, TM, NPP, NPB, BY, HI, and SY show that these parameters can be used as effective selection indices to select genotypes with high yields (Figure 4). The pleiotropic gene effects are probably responsible for the correlations among the quantitative traits [64,66].

Excessive genotypic variation for plant height (PH), number of branches per plant (NBP), leaf area per plant (LAP), and reproductive organs, like number of pods per plant (NPP) and number of seeds per pod (NSP), was evident. They are the major determinants of yield potential since they have a role in determining canopy structure, light interception, and assimilate partitioning [67,68].

The tallest genotypes (GER, PTO, and THEm) were predominantly landraces or large-seeded cultivars, while the small-seed landrace CHR scored the lowest, indicating diverse morphological adaptations. Landraces also indicated strongly elevated branching, which resulted in more pod numbers and, eventually, seed production. Such correlations between branching, number of pods, and yield have also been reported by several researchers [44,62,63].

While the large-seeded genotypes had greater leaf area per plant, this was not necessarily followed by increased yield, suggesting that maximum vegetative growth was not necessarily going to result in enhanced reproductive yield [69]. However, small-seeded landraces LAX and DIG reconciled high pod number and seed number per plant with mid-level canopy size to optimize their source–sink relations optimally and stabilize their yield. The absence of a notable year effect on seed yield per plant (SYP) also testifies to the flexibility of these landraces under diverse climatic conditions [70].

Previous studies have shown that seed physical characteristics, including the thousand seed weight (TSW) and seed diameter (SD), were highly dependent on genotype and could easily differentiate between the small- and large-seeded lines [11,63]. The variation in TSW (18.4–56.5 g) parameter in this study corresponds to the variation reported for lentil germplasm according to the literature [2,5,71]. Even though the large-seeded types like IKAm and THEm had the biggest seeds, the latter had lower per-unit-area seed yield than small-seeded landraces, showing a general trade-off between seed weight and per-unit-area yield [63,72].

Seed protein content (SPC) also differed markedly among genotypes, with small-seeded landraces having higher protein content (maximally 23.35%) than high-seeded cultivars. This trend is consistent with earlier results, demonstrating that small seeds would generally have a higher concentration of protein by dint of a dilution effect arising from greater carbohydrate accumulation in large seeds [49,58]. Moreover, there were also environmental effects on SPC parameter, with an increase in protein content in 2021/2022 due to relatively lower temperatures at the seed filling stage, which improved nitrogen assimilation and limited starch deposition [58].

An overall description of parameter correlations and genotype grouping was provided by the principal component analysis (PCA) (Figure 3). The first two principal components explained total variability at 46.5%, with PC1 separating landraces and small-seeded genotypes (positive axis) from cultivars and large-seeded genotypes (negative axis). Landrace and small-seeded types were strongly correlated with the parameters TF, NP, LAP, PH, NBP, and NPP, testifying to the high adaptability of these genotypes and high agronomic values. On the other hand, cultivars and large-seeded types correlated with seed size traits (TSW and SD), testifying to the priority of market traits over yield stability or plasticity when it comes to breeding. The placement of high-yielding landraces (LAX and DIG) in the quadrant that correlates with SY, SYP, NPP, NSP, and HI bears witness to their excellent performance and can be used as parents in hybridization schemes to tap high yield combined with desirable seed attributes. Such conclusions coincide with earlier multivariate studies in lentil germplasm [11,63].

Overall, the findings present a significant genetic variation between lentil genotypes in regard to primary phenological, morphological, and yield characteristics. The better yield stability, reproductive effectiveness, and seed protein content of small-seeded landraces make them an interesting source of genetic material for breeding purposes. Large-seeded cultivars possess, on the other hand, the market-preferred seed traits, implying that future breeding programs should focus on merging the opposing traits. Genotypes by environment interactions and trait relationships indicated by PCA are valuable for selection strategies in Mediterranean-type environments. Reciprocal crosses between high-yielding small-seeded landraces (e.g., LAX, DIG) and large-seeded cultivars (e.g., IKAm, THEm) can produce recombinants that are more productive and characterized by earliness, as well as acceptable seed size and quality [54,55]. In addition, the stability and endurance of landraces are valuable for conservation and utilization of breeding tolerance to climate variation and for sustainable lentil production systems [54].

5. Conclusions

In conclusion, substantial genetic variation was found among Greek lentil genotypes, for both commercial cultivars and landraces of small- and large-seeded genotypes. High variation was evident for all phenological, morphological, yield, and quality characteristics, reflecting the broad adaptability in this germplasm. Genotype-by-environment interactions were acute, and climatic factors, particularly temperature and rainfall, played a dominant role in affecting performance. Small-seeded landraces exhibited greater seed yield, biological yield, harvest index, and protein content, and greater stability in yield and adaptability to Mediterranean environmental conditions. These landraces were more tolerant of the unstable and stressful environment associated with Mediterranean-type climates, characterized by drought and heat stress, which could potentially worsen as a result of climate change. A high positive relationship between seed yield and traits such as plant height, number of branches and pods per plant, biological yield, and harvest index makes them excellent indirect selection descriptors. Small-seeded genotypes and landraces were separated by principal component analysis as more productive and adaptive, whereas large-seeded cultivars were related to seed size characteristics. In general, Greek lentil landraces could be a treasure trove of genetic material for breeding programs, and the introduction of high-yielding small-seeded landraces (e.g., LAX, DIG) crossed with large-seeded varieties (e.g., IKAm, THEm) could produce new, high-yielding varieties, balancing adaptability, productivity, and market-preferred seed features in favor of sustainable Mediterranean agriculture. However, further research is required to fully understand the effects of abiotic stresses, especially drought and heat, on genotype performance. It will be important in future breeding programs to incorporate genotype-by-environment interactions with precise climate data in order to identify suitable genotypes for Mediterranean-type climates. This is particularly necessary because it has been noted that early-maturing genotypes, especially small-seeded landraces, have shown potential for combating the effects of terminal heat and drought stress, which are expected to increase as a result of climate change. Therefore, early-maturing, high-yielding, and stress-tolerant genotypes should be prioritized in future breeding programs to ensure that lentil production remains viable and productive in diverse and changing climatic conditions.