THE RECOGNITION OF SUBSTANCES, ETC. 375 



tion, we shall refer frequently to the previous statements of the 

 facts. To understand what is here said, therefore, the reader 

 should look up every reference and reread the statement. 



Is the Material a Mixture? The first question that occurs 

 to us is, whether the material is a single substance or a mixture. 

 When 'it is a mixture, we can often, though not always, very 

 quickly find this out. 



If it is a solid mass or powder, we examine it with the naked 

 eye and with the help of a lens. If we see two or more kinds of 

 particles, as in granite (p. 4), in a mixture of sand and sugar, or 

 in a piece of rusty iron, the kinds differing in form or color or both, 

 then it probably is a mixture. 



Whether it appears to the eye to be a mixture, or not, we can 

 next try a solvent, such as water, ether, or carbon tetrachloride. 

 If a part of a small sample refuses to dissolve (e.g. sand), while 

 the rest dissolves (e.g. sugar), we have shown that there are two 

 different sets of physical properties, and therefore two different 

 substances (p. 13) present. As there may be very little of the 

 soluble substance in the mixture, we may not perceive at once that 

 anything has dissolved. So we allow a drop of the liquid to 

 evaporate on a clean watch crystal, and observe whether any resi- 

 due remains. 



If the material is a liquid, we depend largely on differences in 

 the volatility of different substances to find out whether it is 

 a mixture. If a drop evaporates on a watch crystal, leaving a 

 residue (solid or liquid) which does not evaporate, then it is a 

 mixture. If this test fails, because all, or none, evaporates, then 

 we must distil the liquid, with the bulb and most of the stem of a 

 thermometer in the vapor (Fig. 35, p. 67), and note whether the 

 whole comes over at one temperature (single substance, in most 

 cases) or whether the temperature changes as the distillation 

 proceeds (mixture, such as petroleum, p. 344). 



These are simply a few examples intended to show how, when 



