determining the Hardness of Minerals. 409 



of this increment for different diamond points is now being 

 investigated. In a future publication the calibration of the 

 instrument will be described in detail ; in the present work 

 we shall describe only a preliminary series of tests with the 

 minerals of the Mohs scale, in order to show the efficacy of the 

 method. 



The Mohs scale / sclerometric values. — For these tests the 

 diamond point was adjusted slightly out of center, so that it 

 would abrade away a perfect ring-shaped groove and thus avoid 

 the clogging of the hole noticed in Pfaff's experiments. The 

 micrometer scale m was arranged parallel to a cross-hair set in 

 the 45° position from lower left quadrant to upper right, and 

 the inclination of the scale was adjusted so that the two scale 

 divisions nearest the center of the cross-hairs should record a 

 difference of focal depth of 10/>t. It is important that these 

 focal measurements be read always on the same part of the 

 field of the ocular, as there is considerable variation in dif- 

 ferent parts due to aberration. It is also necessary to adopt a 

 uniform criterion of focal perfection ; for the author a fine 

 granular structure observed in the micrometer glass at the side 

 of each scale division afforded a very sharp determinant, accu- 

 rate in every case to a fraction of the value -001 mm or 1/jl. The 

 rate adopted for this series of tests was 6-7 revolutions of the 

 diamond per second, regulated by the governor on the clock- 

 work and by the tension of the belts; this was checked at 

 various times throughout the series and was found to maintain 

 a very uniform average ; the clockwork was wound up to its 

 maximum tension at the beginning of each test, and in the 

 case of the harder minerals at the end of every ten minutes, 

 this being determined as the period through which the main- 

 spring would maintain a uniform rate without appreciable 

 change. The weight used was ten grams, which was found 

 too high for the soft minerals and too low for the very hard 

 ones, snowing the advisability of using two sets and reducing 

 to a common unit as in Pfaff's tests : the constant weight was 

 retained, however, in the following series, in order to discover 

 the possibilities of the instrument, and no great value is placed 

 upon these preliminary results. Many inaccuracies will be 

 discovered in the results attained, especially in the case of the 

 soft minerals, which are averaged in each case from three 

 observations, and these show considerable diversity for the 

 same mineral, depending upon variations in surface texture, 

 and irregularities in the rotation induced by too great weight 

 and irregular resistance, as will be mentioned. The indices i 

 and i x were read before each test and the number of complete 

 revolutions of i x was recorded during the experiment. A 

 preliminary test with soft glass showed the radius of eccen- 



