om 



SCIENCE. 



[N. S. Vol. XIX. No. 484. 



cept in perfecting the apparatus used in 

 making chemical aualj'ses. It is of the 

 utmost importance that their analyses are 

 accurate and quickly performed. On very 

 important work, such as analyses made for 

 settlements on raw material and finished 

 products, analyses are run in duplicate and 

 settlements made on a split between the 

 buyer's and seller's results. This com- 

 petition encourages very accurate work on 

 the parts of the analysts, and they become 

 very skilful. 



It is the custom in all well-managed in- 

 dustrial laboratories to investigate fre- 

 quently the analytical methods used, in or- 

 der to determine their accuracy, reliability, 

 ease and quickness of performance. Old 

 methods of analysis are thus improved, new 

 methods invented and the new methods of 

 others compared and adopted, if found the 

 most suitable. For this reason it is not 

 uncommon to find the most desirable ana- 

 lytical methods iTsed in the laboratories of 

 our important industrial establishments. 

 The colleges would do well to look into these 

 methods as far as possible, and thus keep 

 abreast with the best practice to aid them 

 in teaching analytical methods. 



There is no reason why the training of 

 analysts in large laboratories should not 

 be of the highest order. It is a great mis- 

 take to allow the standing of the work done 

 in these laboratories to run down. It is a 

 grave error to economize too much in the 

 laboratory by employing too few analysts 

 and thus prevent the practical managers 

 from receiving all the information required 

 to control intelligently the various pro- 

 cesses in the factory. 



After men have been a few years in an 

 industrial laboratory they, as a riile, desire 

 positions in the works. It is the exception 

 when we find a chemist from college who 

 has studied mechanical engineering; for 

 this reason only very few chemists become 

 good candidates, qualified for giving proper 



attention to large factory processes where 

 the many complicated devices require en- 

 gineering as well as chemical skill. I have 

 known many of the men in a laboratory to 

 study mechanical engineering either at 

 night schools or with correspondence 

 schools. It would have appeared the part 

 of wisdom for such men to have taken a 

 course at college in mechanical engineering 

 as well as in chemistry, thus fitting them 

 for a wider field of work in their chosen 

 vocation, and affording an opportunity to 

 make greater advancements. 



The future success of any well-estab- 

 lished industrial institution of a chemical 

 nature is in grave peril if it does not have 

 an investigation or research department. 

 The manager of this department must be 

 by education a chemical engineer. He 

 should have had much experience as a 

 practical business manager of plants, and 

 a direct acquaintance in the designing, re- 

 construction and repairs of the same. 

 This department must have a properly 

 equipped research laboratory. The head 

 of this research laboratory must be pos- 

 sessed of very high attainments as a 

 chemist and physicist, with a fair knowl- 

 edge of mechanical engineering. His work 

 through life will be stamped with the great- 

 est success if he has been trained at college 

 in methodical methods of thinking, as ap- 

 plied to original work, and to many ex- 

 amples of practical investigation and ex- 

 perimentation. The chemists under him 

 should have received the same education 

 and training at college. It is desirable 

 that this department should have the capac- 

 ity to investigate new processes that are 

 presented, and if they look promising, a 

 small working plant should be constructed 

 and operated by them to prove fully the 

 value of the method and to give the neces- 

 sary practical data to be used in the designs 

 of a large and fully equipped plant. This 

 department will keep in touch with every- 



