Work on large samples of southern African lithics, probably used to tip hunting weapons amongst other things, exposed limitations in the tip cross-sectional area method's robustness for hypothesising about variation in Stone Age/Palaeolithic weapon-delivery systems. Here we list some of the limitations and discuss a few recently published improvements in tip cross-sectional area ranges and data presentation.Our main contribution is the meaningful enlargement of datasets obtained from hafted weapon tips and/or weapon tips of known use mostly from sub-Saharan Africa. We brie y discuss why this region is relevant for studying trends in the evolution and development of hunting weapons. Our new data are used to strengthen and constrain the different TCSA ranges used as proxies for poisoned arrow tips, unpoisoned arrow tips, javelin tips, stabbing-spear tips, and to suggest a working TCSA range for thrustingspear tips. We demonstrate that the calibrated TCSA ranges have robust statistical integrity as proxies for the different weapon-delivery systems they represent. Apart from the dart-tip category, the TCSA method has now been improved to accommodate more nuanced and accurate interpretations, while further strengthening hypothesis building about ancient weapon systems.