Apeil 20, 1900.] 



SCm'NCE. 



603 



stances shading off into each other by in- 

 finitesimal differences. Why, for example, 

 should there not exist a sodium chloride 

 with 34 or 36 parts chlorine to 23 parts 

 sodium, as well with 35.4 parts? It is true 

 that we may bring these elements together 

 in any proportion, but unless the ratio is just 

 23 to 35.4, the excess of the one or the other 

 will be left unchanged, and we always ob- 

 tain a chloride with 23 parts sodium and 

 35.4 parts chlorine. The hypothesis of con- 

 tinuity does not explain why, in the series 

 of molybdenum chlorides just mentioned, 

 the weights of chlorine combining with a 

 given weight of molybdenum should be in 

 the proportion 2, 3, 4, 5, without any or 

 every intermediate figure. It is quite as 

 difiQcult to explain on this view why the 

 same combining number always adheres to 

 the same element no matter into what com- 

 bine it enters; the combining number of 

 chlorine might be 35.4 with respect to so- 

 dium, and any other figure with regard to 

 lithium or potassium. 



If, however, we assume that a given por- 

 tion of each of the elements, instead of 

 being capable of any degree of subdivision 

 whatever, consists of minute parts, or atoms, 

 each of which, while it may or may not be 

 further divisible, nevertheless always acts 

 in chemical reactions as if it were not, that 

 is, acts as a whole ;• and if we assume that 

 in the same elementary substance, these 

 particles have the same weight, but that 

 the weight differs in the case of each ele- 

 ment, then we have a state of affairs which 

 would necessarily lead to the two laws I have 

 described. The combining numbers would 

 represent simply the relative weights of 

 these chemically ultimate particles ; a so- 

 dium atom weighing 23, would unite with 

 a chlorine atom weighing 35.4, or a bromine 

 atom weighing 80, while a bromine atom 

 weighing 80, would combine with one of 

 potassium weighing 39. So, also, an atom 

 of molybdenum, weighing 96, would unite 



with 2, 3, 4 or 5 chlorine atoms, each weigh- 

 ing 35.4. 



It will be observed that this hypothesis 

 involves no assumption as to the cause or 

 manner of the union of these chemical 

 atoms. Whether they simply lie side by 

 side, each retaining its individuality, or 

 whether they interpenetrate, fuse or blend 

 together, and for the time lose their indi- 

 vidual existence. We shall see presently 

 that there are reasons for adopting the 

 former view. 



The atom, in a chemical sense, may be 

 defined as the smallest portion of an ele- 

 ment which acts as an independent unit in 

 chemical changes ; the chemical molecule is 

 the smallest portion of any substance, ele- 

 mentary or compound, which retains all the 

 chemical properties of the substance in mass 

 and which can move to an unlimited ex-; 

 tent, independently of other portions. The 

 molecules of compounds, therefore, consist 

 of several atoms ; the molecules of ele- 

 ments, there is good reason to believe, are 

 frequently composed of several like atoms, 

 while in other cases they consist of but 

 one. Each kind of atom, therefore, has a 

 specific mass, represented by the combining 

 number, and specific chemical qualities, by 

 virtue of which the elements differ, as iron 

 and sulphur. Of the relation of these little 

 is known, except that the chemical qualities 

 are to some extent periodic functions of the 

 mass. It cannot be asserted that every 

 atom has combining power, for a whole 

 group of elements, the helium-argon group, 

 shows no well-established tendency to form 

 compounds. 



Before proceeding to discuss the fur- 

 ther properties which chemists have been 

 led to attribute to the atoms, we may con- 

 sider certain qualities which have from 

 time to time been ascribed to them, but on 

 which chemistry is silent. Speculators have 

 often erred in attempting to elaborate their 

 hypotheses too fully, and by making as- 



