The purpose of this thesis might be subdivided under two headings - modification of the testing machine so that it would be suitable for drilling tests and development of a standard drilling test for machinability which would employ the modified machine. The first mission was accomplished by the substitution of a specimen-holder in the form of an angle iron for the tool holder already attached to the carriage. One other modification consisted of replacing the timing circuit of the machine with a 1 rpm indicator card mount which, together with a recording pen, gave a pictorial record of the progress of the drill through test specimens. The second purpose of the investigation was fulfilled when the straight line portion of the machinability curve was found, proved, and defined by equation. Also of equal interest with regards to this machinability concept was the establishment of confidence limits showing the relative variability of the materials tested. Perhaps the statement should be made that the exact values of these confidence limits depend on the number of test specimens being included in the index determination. The investigators based their calculations on a sub-group containing four specimens. This sub-group size was believed sufficiently high to detect any test equipment defect, such as a feather edge on the drill or an incorrectly set speed, without requiring an excessive number of drilling operations. Although the index values calculated and listed in Table #13 are correct for the materials tested, they are not recommended for general use since they are based on test specimens all coming from one source. A larger number of tests on specimens from a large number of foundries and steel mills would be necessary for a true estimate of the machinability of the test materials. / M.S.