332 PROCEEDINGS OF SECTION B. 



In the remaining determinations the methods given by Hillebrand, 

 Washington, or Treadwell Hall are followed as closely as possible. 



Instead of analyses being made in duplicate, a satisfactory plan to 

 adopt for safe work is for single analyses to be made by two men, and 

 a comparison made. Any serious discrepancy should be checked by 

 repeating. The figures finally chosen, should, however, be those of one 

 analyst — at any rate for each group of determinations. 



The variations suggested in this paper have been arrived at only 

 after a long series of analyses worked as closely as possible on the lines 

 laid down by the authorities quoted. No one can deny but that the 

 care and exactness with which a master such as Hillebrand has arrived 

 at his conclusions are worth the closest consideration of any chemist 

 desirous of turning out the best work in the analysis of rocks. The 

 modifications suggested are intended to simplify some of the operations 

 involved, and it is thought that the results obtained are quite within 

 the limits of experimental error. 



It is much to be desired that an organised classification of the 

 igneous rocks of Australia should be instituted on the lines adopted by 

 the Geological Survey of U.S.A., and this could be done by the individual 

 chemists of each State, if opportunity were made. But it is surely 

 necessary at the outset in this, as in any independent work which is 

 intended for comparison with other similar work, that a uniformity of 

 method should be adopted which would render such a comparison of 

 the greatest real value. 



In the closer sympathv and co-operation which, with the develop- 

 ment of Commonwealth ideals, must surely come into scientific work 

 earliest of any, the agreement upon a standard plan of action is surely 

 the first thing to be desired. A Congress such as this, with the record 

 of those that have preceded it, is the best guarantee that such an 

 agreement between analysts in each State would be heartily welcomed. 

 Even if it were not convenient for any particular State to take up its 

 own share at once, it would always be ready to hand, and, under standard 

 conditions, the work of any accredited analyst would always be accept- 

 able for inclusion in the general scheme. In the Geological Survey of 

 Victoria the organised analysis of the igneous rocks of that State has 

 been in operation for several years, and the results are shortly to be 

 issued as a bulletin. Unfortunately, it has not been possible to set 

 apart a definite portion of the staff for this work. This is the most 

 satisfactory manner of conducting such an examination. 



I would like here to acknowledge the valuable help and co-opera- 

 tion afforded by Mr. Alan G. Hall, of the Mines Department laboratory, 

 in the preparation of experiments for this paper. 



In the ultimate collection and arrangement of the results from 

 different sources much uncertainty would be avoided if the exact 

 value of available analyses could be determined, and this is only 

 possible if the methods adopted and precaution observed are those 

 upon which an agreement has been arrived at and a satisfactory 

 standard fixed. 



