Wheat landraces are a very valuable genetic resource for the different contemporary cereal-based farming systems. Camacho Villa and collaborators proposed the following definition: “a landrace is a dynamic population of a cultivated plant that has historical origin, distinct identity and lacks formal crop improvement, as well as often being genetically diverse, locally adapted and associated with traditional farming systems” (Villa et al. 2005). Until the end of the nineteenth century landraces were the principal resource of agricultural production (Harlan 1975). A gradual replacement during the early decades of the 20th century by selected component pure lines and modern cultivars happened. Nevertheless, the persistence of landraces in different environments was due to their increased stability, accomplished through generations of natural and deliberate selection for valuable genes for resistance to biotic and abiotic stresses and intergenotypic competition, as well as to their favorable morpho-physiological and agronomic traits tightly linked with the high yield ability and quality of products. Besides, landraces represent the maximum expression of the adaptation to the specific environmental conditions and agronomic management of the areas were they evolved (Moragues et al. 2007). There are a number of wild species, landraces, and traditional cultivars within the Triticum genus that constitute the wheats of the world. The main center of diversity is the Fertile Crescent, extending from the Mediterranean coast in the west to the east. In this region, diploid and polyploid Triticum species coexist in mixed populations and exhibit a wide range of morphological and ecological diversity (Hoisington et al. 1999). Among polyploid species, tetraploid wheats (Triticum turgidum L.) belong to a taxonomic category that includes genetically and morphologically different entity, and their evolution under domestication has not been fully explained. Archaeological findings and genetic studies indicate that emmer [Triticum turgidum L. subsp. dicoccum (Schrank) Thell.], the first domesticated form of tetraploid hulled wheat originated from the tetraploid wild ancestor in the western half of the Fertile Crescent 7,800-7,500 before present. Free-threshing tetraploid naked wheats, rivet (Triticum turgidum L. subsp. turgidum) also called poulard, cone or english wheat, and afterwards durum [Triticum turgidum L. subsp. durum (Desf.) Husn.], evolved from emmer in the Near East and spread through the north side of the Mediterranean area reaching the Iberian Peninsula and Algeria from Italy. It has also been hypothesized that the rivet wheat arrived in England with the Normans from Sicily, and according to Percival (Percival 1921) surely there are a group of rivet landraces adapted only for growing in countries bordering the Mediterranean. Through the evolution in mountain environments, rivet wheat has acquired rusticity in terms of could tolerance, ability to grow in marginal soils, weed competitiveness and resistance to diseases. Moreover, due to the kernel hardness not fully vitreousness as durum wheat, the grain of rivet wheat has been traditionally used preferably for bread making or for homemade pasta and biscuit preparation. However, nowadays, durum wheat remains the most important cultivated tetraploid wheat in the Mediterranean basin, Canada and USA. In the Mediterranean region, it is mainly grown in the Middle and Near East regions and North Africa. Based on cytological and molecular analyses, Triticum turgidum is believed to be originated from a hybridization between Triticum urartu (AA genome, n = 7) and an unknown diploid species (BB genome, n = 7) closely related to Aegilops speltoides (Dubcovsky and Dvorak 2007). Thus, tetraploid wheats as rivet and durum wheat, are a self-pollinated allotetraploid cereal (harboring two genomes with genomic formula: AABB) with 28 chromosomes (2n = 4x = 28). In addition to the taxonomical classifications proposed by both Van Slageren and MacKey, which consider rivet and durum wheats as subspecies of T. turgidum, MacKey further classified these forms as covarieties of the subspecies T. turgidum subsp. turgidum (T. turgidum subsp. turgidum conv. durum), whereas Dorofeev and colleagues considered tetraploid wheats as individual species T. durum. Nevertheless, the more recent results obtained by Oliveira and collaborators (Oliveira et al. 2012) on the basis different marker types (nuSSRs, ISBPs, cp-haplotypes) and several analysis methods (STRUCTURE, PCA, genetic distance approaches), support MacKey’s taxonomical classification, which consider rivet and durum wheats as varieties of the same taxon. The authors also suggest that rivet landraces constitute an easily transferable source of genetic variation for durum wheat breeding programs. Fortunately, although rivet has been neglected and it disappeared from cultivation during the last century, accessions have been preserved by the inclusion in germplasm bank collections and are available for new breeding activity. The knowledge of the extent and pattern of genetic diversity within and among wheat populations is a key factor for the identification of useful genotypes for plant breeding purposes and to better understand the crop requirements to design appropriate collection and conservation strategies (Asmamaw et al. 2019). Furthermore, a renewed focus on wheat landraces for breeding purposes could relieve some negative consequences of intensive agriculture and conventional breeding, such as the irrational and/or excessive use of auxiliary input, excessive homogeneity of cropping systems, the loss of genetic diversity (Tilman 1998), and the stagnation of yields in marginal cereal areas (Annicchiarico and Pecetti 1998). This is also functional for the definition of a plant ideotype suitable for low-input farming systems, mainly smallholder and organic (Anastasi et al.). In Italy, tetraploid wheats, especially durum wheat, have a long-time tradition of growing and breeding, and accessions collected in Southern Italy, which include rivet germplasm, now preserved ex situ, are a valuable genetic resource. A large number of Italian and some North African landraces of tetraploid wheats were grown in Italy since the end 1800 and after 1920 were gradually substituted by the pure line varieties selected from Italian landraces, but also from Syria-Palestinia and North Africa (Lybia) germplasm. Starting from the second half of the last century, also the internal hilly areas of Sicily characterized by the cultivation of wheat local populations, including some exaploid wheat (Triticum aestivum aestivum), have drastically contracted until almost total disappearance due to the advent of an increasing number of novel improved varieties, which are able to exploit grater levels of auxiliary input, particularly for nitrogen nutrition and weed control. In the mountain environments, this phenomenon occurred much more slowly because the so-called “modern” varieties hardly adapted to the extreme variability of local pedoclimatic conditions. The evolution of these landraces has been characterized not only by the particular pedoclimatic conditions of the Sicilian mountain areas, but also from the socio-economic and productive needs that this territory imposes. The selection exercised by the farmer-breeder on these local populations over the centuries has mainly taken into account two fundamental aspects: the production of semolina, suitable for processing into large-sized traditional breads (from 2 to 4 kg) suitable for long-term shelf-life during transhumance, and the high production of straw essential in cattle farms. The considerable advances in molecular genotyping and databasing technologies in recent years are beginning to make the variation and resources of landraces more accessible for exploitation. High-throughput genotyping enables Genbank accessions with uncertain provenance to be elucidated and thereby enables validation of associated phenotypic data, making them much more useful (Newton et al. 2009). Molecular markers, such as RFLP, SSR and SNP have been successfully used for identification of cultivars, diversity estimates, and genetic relationships assessment in crops including rivet and durum wheat (Autrique et al. 1996; Maccaferri et al. 2003; Kabbaj et al. 2017; Fiore et al. 2019). For their high polymorphism, codominance and locus specificity, simple sequence repeats or microsatellite (SSRs) markers have proved to be highly efficient molecular tools for the characterization of durum wheat germplasm collections (Maccaferri et al. 2003; Moragues et al. 2007; Figliuolo et al. 2007; Ruiz et al. 2012; Laidò et al. 2013; Sahri et al. 2014). Different authors in the past years developed physical consensus maps of SSR markers in both soft wheat and durum wheat chromosomes (Röder et al. 1998; Somers et al. 2004; Paux et al. 2012). To date most studies on Italian durum germplasm have analyzed collections including old and new elite varieties for morphophysiological and qualitative traits (De Vita et al. 2007; Motzo and Giunta 2007), and the use of molecular markers was focused on temporal trends of diversity (Hoisington et al. 1999; Giunta et al. 2007; Kabbaj et al. 2017), relatedness among genotypes (Maccaferri et al. 2003), genetic structure (Martos et al. 2005), also in comparison to Triticum turgidum L. subspecies (Laidò et al. 2013). In a recent study (Kabbaj et al. 2017) a panel of 370 durum wheat genotypes including 35 Italian genotypes were genotyped using 500 single nucleotide polymorphism (SNPs) markers. In 2018, Marzario and colleagues (Marzario et al. 2018) used a smaller number (44) of simple sequence repeats (SSR) molecular markers to detect pattern of diversity for 136 accessions collected in the South Italy over time, to identify the genepool of origin, and establish similarities with 28 Italian varieties with known pedigree grown in Italy over the same time-period. They also conducted phenotyping for 12 morphophysiological traits of agronomic interest thus obtaining enough information on the genetic structure of durum wheat genotypes for a quick screening of the germplasm collection. More recently, in 2019, Asmamaw and collaborators (Asmamaw et al. 2019) assessed the magnitude and pattern of genetic diversity in Ethiopian durum wheat landraces by SSR molecular markers analysis. Furthermore, Fiore and collaborators in 2019 (Fiore et al. 2019) characterized a collection of durum wheat landraces from Sicily, using single nucleotide polymorphisms (SNP) markers, together with agro-morphological, phenological and quality-related traits. For the above-mentioned reasons, in this study a strategy combining molecular genotyping and morpho-agronomic traits is proposed to characterize a group of rivet wheats named “Bufala”, historically cultivated in the mountain areas of Sicily (from 800 to 1,200 m a.s.l.) and used for the production of a traditional bread type locally appreciated. In particular, a total of 55 SSR molecular markers were utilized to detect pattern of diversity for 30 tetraploid wheat genotypes, a collection of twenty “Bufala” and seven “Bufala-related” rivet landraces, in comparison with an outgroup of three improved durum wheat varieties., Furthermore, phenotyping was than conducted for a set of significant morpho-agronomic traits, potentially useful for breeding purposes to obtain new genotypes suitable for low-input farming systems.