In the studies of aked stone tools, the "quality" of lithic raw material has often been assessed from a viewpoint of its suitability to lithic production, and several studies quanti ed such raw material qualities by measuring mechanical properties of several different rocks like obsidian, chert, and basalt. To advance this approach, this study aims to quantify variations in the quality of chert, which is usually based on the difference in grain-size, e.g., ne-grained chert as "high-quality". For this purpose, we measured two types of hardness, i.e., the rebound hardness by Schmidt Hammer and the Rockwell Hardness (HRC), by using chert samples from southern Jordan, where the use of ne-grained chert increased in the Early Upper Paleolithic (Ahmarian) in association with the development of bladelet technology. The results indicate that ne-grained chert has greater fracture predictability and particularly suitable for the production of bladelets. However, ne-grained chert in southern Jordan suffer from abundant internal fractures and was not used as often as medium-grained chert for Levallois products and robust blades in the Late Middle and Initial Upper Paleolithic. Thus, explaining the prehistoric selectivity of lithic raw materials need to consider fracture predictability as well as other factors, such as actual conditions of occurrences and lithic production technology.