Abstract: Garlic (Allium sativum L.) is one of the most important bulb vegetables grown and used as spice and flavoring agent for foods. It has benefits in lowering total plasma cholesterol, reducing blood pressure and decreasing platelet aggregation. The center of origin of garlic has been considered to be central Asia. From the center of origin, garlic has been spread to west, south and east. In Iran cultivation and consumption of garlic has a long history and areas under its cultivation is estimated about 10 000 ha. Garlic is a sterile species and reproduces only by vegetative propagation. Although garlic is an asexually propagated crop, it displays great morphological diversity in bulb, and leaf size, color and shape, scape presence and height, and flower color, fertility and bulbil (topset) development. Until recently, characterization of garlic germplasm has been based primarily on morphological data. However, morphological characters may differ under varying environmental conditions. Many classification systems for garlic have been described. Using molecular techniques scientists are able to evaluate diversity between ecotypes. Isozyme and RAPD markers have been utilized to classify and categorize genetic diversity in garlic. The random amplified polymorphic DNA (RAPD) technique provides high numbers of markers which can be used for genetic studies. RAPD have the advantage that the material is processed by an efficient and inexpensive technique without requiring prior knowledge of the genome. The purpose of this study was to detect molecular polymorphism among Iranian garlic by RAPD. We accomplished this by using total genomic extracts from diverse garlic clones, collected from the main cultivation areas of Iran. These will help to delimit varietals groups in Iranian garlic ecotypes. Genomic DNA was extracted from young leaves according to the Dellaporta method. PCR was carried out using ten 10-mers arbitrary primers. Results indicated that five of them had no amplification. A total of 22 RAPD bands have been produced. The similarity matrices and dendrogram were obtained using UPGMA algorithms. In conclusion the entire were grouped into four clusters with 6, 3, and 9 which indicate there is a significant difference between Iranian garlic accessions.