312 PERSONAL PRiifAKAl'iON FOR RESEARCH. 



together. In nearly all chemical experiments, in order to 

 prevent interferences, the first preliminary condition to be 

 secured is a high degree of purity of the substances. Even 

 when a new or uninvestigated phenomenon does really 

 exist, the exclusion of interferences takes a long time, and 

 then another long period is consumed in developing the 

 phenomenon to a degree of magnitude suitable for investi- 

 gation ; and very few of the experiments made up to this 

 point can be published, because they are imperfect. 



In actual research, the experiment or apparatus is 

 often arranged in such a way that the order and disposi- 

 tion of its different parts can be readily altered ; and 

 whilst the experiments are in progress, the direction, dis- 

 tance, and strength of the forces employed are varied, 

 instead of preparing afresh for every modified trial. By 

 this plan a number of results and the means of drawing 

 many conclusions are obtained with the trouble of only 

 one preparation. For example, if a large voltaic battery 

 is employed, a number of experiments are prepared before 

 charging it, and the battery is so arranged that the mag- 

 nitude and tension of its current can be easily varied. 



The number and extent of the preparations necessary 

 for experiments vary, of course, with the character and 

 magnitude of the research. Nearly always, the amount of 

 trouble required to prepare for experiments is very much 

 greater than that expended upon the actual trials. Wait- 

 ing for substances and apparatus is also a frequent source 

 of delay, and clearing away the residues of experiments 

 consumes a great deal of time. 



For the purposes of experiment, every physical investi- 

 gator requires a suitable and sufficient supply of materials 

 and apparatus, especially those necessary for generating, 

 governing, detecting, and measuring the different forces 

 of nature. Every chemical experimentalist also needs 



