April 20, 1900.] 



^yCIENCE. 



605 



energy manifests itself, of like properties in 

 the same element, but differing in mass and 

 certain other respects in the different ele- 

 ments. 



We may now consider some of the prop- 

 erties which the chemist does ascribe to the 

 atom in addition to mass and specific chem- 

 ical nature. 



In recent years much has been learned 

 about chemical affinity, a great part of which 

 is susceptible of mathematical expression, 

 and is independent of the atomic theory, 

 only the experimentally determined com- 

 bining numbers coming into consideration, 

 the gram atom and gram molecule taking the 

 place of the atom and molecule. As to the 

 real cause or nature of affinity we are still 

 totally in the dark. Very elaborate re- 

 searches, however, have brought to light 

 certain important facts to which I may 

 properly refer, as they are at present incap- 

 able of interpretation apart from the hy- 

 pothesis of atoms. 



Each unit of an element is capable of 

 uniting with but a limited number of other 

 units, which differs in different cases. This 

 may be illustrated by the hydrogen com- 

 pounds 



CIH, 0H„ NH3, CH,. 



The carbon atom can hold four hydrogen 

 atoms, the nitrogen atom but three, the 

 oxygen atom but two, and the chlorine atom 

 but one. The combining power of carbon, 

 nitrogen and oxygen atoms may be regarded 

 as divisible into four, three and two parts 

 respectively (with regard to hydrogen) 

 while that of the chlorine atom is indivisi- 

 ble. The number of parts into which the 

 combining power is thus divisible is termed 

 the valency of the element, and each of these 

 parts is termed a valence, bond, or affinity unit. 

 Without going further into detail (for the 

 subject is an elaborate one), it may be 

 stated as a general law, that combination 

 takes place by a valence of one atom acting 



on a valence of another atom, or by the 

 several valences of one, acting on the corre- 

 sponding number of valences of another 

 atom or atoms. This is conveniently repre- 

 sented by lines joining the atomic symbols. 

 Thus CH^ and H^O may be represented by 

 the formulas 

 H 



I 

 H— C — H, and H — O — H 



I 

 H 



as the oxygen atom has two valences, these 

 may combine each with a valence of car- 

 bon, thus 



o = c = o. 



This law holds invariably in the case 

 of carbon compounds, and in general, but 

 whether combination takes place only in 

 this way in all classes of compounds is as 

 yet an open question. The number of va- 

 lences has nothing to do with the strength 

 of affinity ; a pentavalent atom has not five 

 times the affinity of a univalent atom. 



It was long ago discovered that there could 

 exist several substances of the same per- 

 centage composition and molecular weight, 

 but differing in chemical and physical prop- 

 erties, the so-called isomeric compounds. 

 Sometimes as many as twenty distinct com- 

 pounds of the same composition are known. 

 This difference is inconceivable if the atoms 

 are indiscriminately arranged, like a lot of 

 different colored balls thrown together at 

 random ; there must be in each a definite 

 arrangement of the atoms which cannot be 

 changed without changing the nature of the 

 compound. A comprehensive study of com- 

 pounds, aided by the conception of val- 

 ency, has led to the idea of the linkage of 

 atoms in the molecule and to the so-called 

 structural or constitutional formulas. The 

 structural formulas of the two forms of 

 butane, C^Hj„, are given, as illustrating the 

 linkage of atoms, as well as the nature of 

 isomerism. 



