388 MEETING OF THE BRITISH ASSOCIATION. 



as moD-atomic radicles, was found to satisfy the requirements of the 

 synoptic formulas. Gerhardt was the first to adopt them from us 

 He gave, in his admirable 'Traite de Chimie Organique,' a system 

 of organic chemistry on that plan ; and his book has been of im- 

 mense service to the development of our science. The extension of 

 these principles to mineral chemistry had been commenced in the 

 cases of the commonest acids and bases ; but their general intro- 

 duction met with difficulties, and sometimes seemed wanting to 

 their complete success. I must now travel southward for a short 

 time, and ask you to accompany me to the sunny land of glorious 

 memories, and to its southern dependency — the Island of Sicily. It 

 was reserved for Professor Cannizzaro, of the University of Paler- 

 mo, to show us how the remainder of the knot could be untied. 

 He argued, upon physical as well as chemical grounds, that the 

 atomic weight of many metals ought to be doubled, as well as 

 those of oxygen, sulphur, and carbon. His conclusion is confirmed 

 by the constitution of those organo-metallic bodies which I mention- 

 ed just now ; and it certainly does seem to supply what was still 

 wanting for the non-metallic elements to the heavy metals them- 

 selves. Theelements are now arranged into two principal groups : — 

 1st. Those of which each atom combines with an uneven number 

 of atoms of chlorine or hydrogen. 2nd. Those of which each 

 atom combines with an even number of atoms of chlorine or hy- 

 drogen. Like every classification founded upon nature, this one 

 draws no absolute line, as some elements belong to both classes.. 

 The first group includes the mon-atomic elements of the chlorine 

 family, the tri-atomic elements of the nitrogen family, hydrogen 

 and the alkali-metals, silver and gold — in all about eighteen ele- 

 ments. The usual atomic weights of these are retained. The 

 usual atomic weights of all the other elements, biatomic, tetratomic 

 &c, are doubled. The second group includes the oxygen family 

 — carbon, silicon, and the alkaline earths ; the metals, zinc, iron, 

 copper, lead, <fec. Every step in our theoretical development of 

 chemistry has served to consolidate and extend the atomic theory ; 

 but it is interesting to observe that the retention of that theory 

 has involved the necessity of deprving it of the absolute character 

 which it at first possessed. Organic compounds were long ago dis- 

 covered, containing atoms of carbon, hydrogen and oxygen, in 

 proportions far from simple ; and the atomic theory must have been 

 abandoned but for the discovery that the atomic, or rather mole- 

 cular weights of these compounds correspond invariably to entire- 



