■Nov. 1 9, 1874] 



NATURE 



53 



donee upon which its composition is b^iscd, and explains, 

 in some cases, how he arrived at the relative weights of 

 the ultimate particles in question. Between the years 

 1S05 and iSlo, however, considerable changes had been 

 made by Dalton in the numbers ; the table found in the 

 first part of the "New System " being not only much more 

 extended, but, in many cases, the numbers differing alto- 

 gether from those given in the first table published in 

 1805. It is therefore now, to a considerable extent, a 

 matter of conjecture how Dalton obtained the first set of 

 numbers ; all we know is that it was mainly by the con- 

 sideration of the composition of certain simple gaseous 

 compounds of the elements that he arrived at his conclu- 

 sions, and in order that we may form some idea of the 

 data he employed, we must make use of the knowledge 

 which chemists at that time (1S03 5) possessed concerning 

 the composition of the more simple compound gases. 



As I can find no record of any explanation of these 

 early numbers, I venture to bring the following attempt 

 to trace their origin before the Society to whom we owe 

 their publication. 



The first point to ascertain, if possible, is how Dalton 

 arrived at the relation between the atomic weights of 

 hydrogen and oxygen given ia the table as I to 5'5 (but 

 altered to i to 7 in iSoS). The composition of water by 

 weii;ht had been ascertained by the experiments of 

 Cavendish and Lavoisier to be represented by the 

 numbers 15 of hydrogen to 85 of oxygen, and this result 

 was generally accepted by chemists at the time, amongst 

 others doubtless by IJalton. Whether in those early days 

 Dalton had actually repeated or confirmed these experi- 

 ments appears improbable. At any rate, he formed the 

 opinion that water was what he called a binary compound, 

 z'.t'., that it is made up of one atom of oxygen and one 

 atom of hydrogen combined togetlier. Hence, if he took 

 the numbers 85 to i 5 as giving the composition of water, 

 the relation of hydrogen to oxygen would be i to 5-6, or 

 nearly that which he adopted. It does not appear possible 

 to explain why Dalton adopted 5-5 instead of 56 for 

 oxygen ; it may, perhaps, have been a mistake, as there 

 are two evident mistakes in the table, viz., 137 for nitrous 

 oxide instead of 139, and 93 for nitrous gas instead of 



9'7- 



Let us next endeavour to ascertain how he obtained the 

 number 4-3 for carbon (altered to 5 in 1808 and to 5-4 

 later on). Lavoisier, in the autumn of 1783, had ascer- 

 tained the composition of carbonic acid gas by heating a 

 given weight ot carbon with ox:de of lead, and he came 

 to the conclusion that this gas contained 28 parts by 

 weight of carbon to 73 parts by weight of oxygen. Now 

 Dalton not only was acquainted with the properties 

 and composition of carbonic acid, but he was aware that 

 Cruikshank had shown in 1800 that the only other known 

 compound of carbon and oxygen, carbonic oxide gas, 

 yields its own bulk of carbonic acid when mixed with 

 oxygen and burnt; and also that Desormes* analysed 

 both these gases, finding carbonic oxide to contain 44 

 of carbon to 56 of oxygen, whilst carbonic acid con- 

 tained to 44 of carbon 1 12 of oxygen, being just double 

 of that in the carbonic oxide. Dalton adds : " This 

 most striking circumstance seems to have wholly escaped 

 their notice." Hence Dalton assumed that one atom of 

 carbon is united in the case of carbonic oxide with one 

 atom of oxygen, whilst carbonic acid possessed the more 

 complicated composition and contains two atoms of 

 oxygen to one of carbon. Now, if carbonic acid contains 

 carbon and oxygen in the proportion of 28 to 72, carbonic 

 oxide must contain half as much oxygen, viz., 28 of 

 carbon to 36 of oxygen ; and assuming that the atomic 

 weight of oxygen is 5-5, that of carbon must be 



Having thus arrived at the number 4-3 as the first 



* Ann. tier Chemie, tome 39, i). 38. 



atomic weight of carbon, it is easy to see why D.ilton 

 gave 6'3 as the atomic weight of carburetted hydrogen 

 from stagnant water, and 5^3 as that of olefiant gas. The 

 one represents one atom of carbon to two of hydrogen, 

 the other one of carbon to one of hydrogen ; or, olefiant 

 gas contains to equal quantities of carbon only half as 

 much hydrogen as marsh gas. This conclusion douVjlless 

 expressed the results of Dalton's own experiments upon 

 these two gases, which were made, as we know from him- 

 self, in the summer of the year 1804. He proved that 

 neither of these gases contained anything besides carbon 

 and hydrogen, and ascertained, by exploding with oxygen 

 in a Volta's eudiometer, that if we reckon the carbon in 

 each the same, then carburetted hydrogen contains 

 exactly twice as much hydrogen as olefiant gas docs, and 

 that ''just half of the oxygen expended on its combustion 

 was applied to the hydrogen, and the other half to the 

 charcoal. This leading fact afforded a clue to its consti- 

 tution." Whereas, in the case of olefiant gas, two parts of 

 oxygen are spent upon the charcoal, and one part upon 

 the hydrogen. 



The atomic weight of nitrogen (azote = 4'2) was doubt- 

 less obtained from the consideration of the composition 

 of ammonia, whose atomic weight is given in the table 

 at 5'2. Ammonia was discovered in 1774 by Priest- 

 ley, but the composition was ascertained by BertlioUet 

 in 1775 by splitting it into its constituent elements 

 by means of electricity, when he came to the con- 

 clusion that it contained 0-193 P^^'s by weight of 

 hydrogen to o 807 parts by weight of nitrogen. Dalton as- 

 sumed that this substance is a compound of one atom of 

 hydrogen with one of nitrogen, and hence he obtained for 



the atomic weight of azote 



807+1 _ 



193 



= 4-2; and 4-2-1-1 = ; 



as the atomic weight of ammonia. It is also probable 

 that Dalton made use of the composition of the oxides of 

 nitrogen tor the purpose of obtaining the atomic weight of 

 nitrogen. If we take the numbers obtained partly by Davy 

 and partly by himself, as given on page 31S of the '• New 

 System," as representing the composition of the three 

 lowest oxides, it appears that the mean value for nitrogen 

 is 43 when oxygen is taken as 5-5. In all probability 

 the number in this table (4-2) was obtained from an expe- 

 riment of Dalton's made at an earlier date. 



It is not possible to ascertain the exact grounds upon 

 which Dalton gave the number 72 for phosphorus ; its 

 juxtaposition, however, in the table, to phosphurette<l 

 hydrogen, shows that it was probably an an ilysis or a 

 density determination of this gas wJiich led hiui to the 

 atomic weight 7-2, under the supposition that this gas 

 (like ammonia) consisted of one atom of each of its com- 

 ponents. In the second table, published in 1808, Dalton 

 gives the number 9 as that of the relative weight of the 

 phosphorus atom, and we are able to trace the origin of 

 this latter number, although that of 7-2 is lost to us. On 

 p. 460, Part 11. of his " New System," Dalton states that 

 he found ico cubic inches of phosphuretted hydrogen to 

 weigh 26 grains, the same bulk of hydrogen weighing 2-5 



grains. Hence ^—7— = 9 S^^^^ '^^ atomic weight ( f 

 phosphorus. It was probably by similar reasoning 

 from a still more inaccurate experiment than this one, 

 that he obtained the number 7-2. 



Sulphur, which stands in the first table of 1803 at 14-4, 

 was altered in thelist published in the" NewSystem" to 13. 

 These numbers were derived from a consideration (l) 01 

 the composition of sulphuretted hydrogen, which he n- 

 garded as a compound of one atom of sulphur with one of 

 hydrogen, and (2) of that of sulphurous acid, which he 

 supposed to contain one atom of sulphur to two of oxygen. 

 Dalton knew that the first of these compounds con- 

 tained its own volume of hydrogen, and he determined its 

 specific gravity, so that by deducting from the weight ot 

 one volume of the gas that of one volume of hydrogen, he 



