lO Sketch of the latest Improvements [Jan, 



combines with two doses of oxygen, and forms two oxides, tlie first 

 composed of J 00 mercury and four oxygen, and tlie second of 100 

 mercury and eight oxygen, it is obvious tiiat tlie first must be a 

 compound of one atom mercury and one atom oxygen, and the 

 second of one atom mercury and two atoms oxygen. 



Nor is there any difficulty with respect to iron. There are two 

 oxides of that metal : the first composed of 100 iron and 28 

 oxygen ; the second of 100 iron and 42 oxygen. Now as 28 is to 

 42 as two to three, it follows that the first is a compound of one 

 atom iron and two atoms oxygen ; the second, of one atom iron 

 and three atoms oxygen. The same rule holds good with respect to 

 the oxides of nickel and cobalt. 



If we know the number of atoms of which a body is combined, 

 and the proportion of the constituents, there is no difiiculty in de- 

 termining tlie proportional weight of the atoms of which it is com- 

 posed. Thus if water be composed of one atom of oxygen and one 

 atom of hydrogen, and if the weight of the oxygen in water is to 

 that of the hydrogen as /-^^ to one, then it follows that the weight 

 of an atom of oxygen is to that of an atom of hydrogen as 7^ to 

 one. If black oxide of mercury be composed of one atom of 

 mercury and one atom of oxygen, and if it be composed of 100 

 mercury and four oxygen, then an atom of mercury is to the weight 

 of an atom of oxygen as 100 to four, or as 25 to one. If black 

 oxide of iron be composed of one atom iron and two atoms 

 oxygen, and if it consist of 100 iron and 28 oxygen, then an atom 

 .of iron is to an atom of oxygen as 100 to 14, or as 7' 142 to one. 

 "Such is the method of determining the weight of an atom of the 

 different substances upon which experiment has hitherto been made. 

 The advantage of such a knowledge is immense; because it gives 

 us the proportions in whicli the ditt'erent substances unite together, 

 and even enables us to calculate the proportional constituents of all 

 compound bodies, independent of experiment, and with more accu- 

 racy than would result from experiments unless conducted with 

 uncommon precautions. 



Hitherto the only persons who have written upon the subject of 

 chemical atoms are Mr. Dalton, Sir Humphry Davy, Dr. JBerze- 

 lius, Dr. WoUaston, and myself. Mr. Dalton made choice of 

 hydrogen as his unit, because it is the lightest of nil the atoms ; 

 and Sir H. Davy has followed his example. But as oxygen enters 

 into a much greater number of compounds than any otiier body, it 

 was chosen by Dr. Wollaston and Dr. Bcrzclius as the most conve- 

 nient unit J and in the tables of atoms which I have published in 

 the different volumes of the Amiak of Philosophj/, I iiave followed 

 their example. Berzcllus considers an atom of oxygen to weigh 

 100, Wollaston makes it weigh 10, and 1 myself make its weight 

 one. The reader will perceive that these three numbers are the 

 same, the only difference being the position of the decimal point. 



The person who has hitherto made the greatest number of expe- 

 riments upon this important subject is Dr. Berzelius ; and be has 



