184 PROFESSOR THOMAS GRAHAM'S SCIENTIFIC WORK. 



class of soluble sub-phosphates analogous to the yellow insoluble phos- 

 phate of silver ; and he showed, with g^eat clearness, that in the three 

 classes of phosphates, namely, the sub-phosphates, the common phos- 

 phates, and the bi-phosphates, the ratio borne to the oxygen of the acid 

 by the other oxygen of the salt is the same, namely, as 3 to 5 5 only that, 

 in the three classes of salts, the non-acid oxygen is divided between 

 different proportions of metallic base and water, thus : 



Sub-phospbate of soda BNaO.POs. 



Coimnon pbospliate o'f soda H . 2 Na . P O 5. 



Bi-phosphate of soda 2HO.NaO.P06. 



He further pointed out that, to these three series of salts, there cor- 

 responded a definite phosphate of water, or, 



Hy dialed phosphoric acid 3 H O . P O .-,■ 



Compounds of one and the same anhydrous acid with one and the 

 same auliydrous base, in different proportions, had long been known ; 

 but it was thus that Mr. Graham first established the notion of poly- 

 basic compounds — the notion of a class of hydrated acids having more 

 than one proportion of water replaceable by metallic oxide, and that 

 successively, so as to furnish more and more basic salts, all preserving, 

 as we should now say, the same type. 



Mr. Graham further showed that Dr. Clark's pyrophosphate of soda, 

 like the common phosphate, yielded an acid-salt or bi-phosphate; and 

 that these two compounds were related to a hydrated phosphoric acid 

 differing in con)i)osition and properties from the above-mentioned hy- 

 drate, and yielding, after neutralization with alkali, a white instead of 

 a yellow prexjipitate with nitrate of silver. This series of compounds 

 he expressed by the following formuhe: 



Claik's pyrophosphate of soda 2 Na O . P O5. 



Acid or bi-pyropliosphate of soda HO.NaO. PO5. 



Hydrated pyrophosphoric acid 2H0. POg. 



Lastly, Mr. Graham showed that when the bi-phosphate or bi-pyro- 

 ])hosphate of soda was ignited, tbere was left a new variety of phos- 

 phate, which he called the metaphosphate, having the same proportions 

 of soda and anhydrous phosphoric acid as the original compound, but 

 differing Irom it in several properties, more parti(,'ularly in its inability 

 to furnish any acid salt. From this new i)hosphate he obtained the cor- 

 responding hydrated acid, and found it to be identical witli that variety 

 of phosphoric acid then, and still, known as glacial phosphoric acid, 

 which had previously been noticed to possess the distiuctive property 

 of causing a precii>itate in solutions of albumen. This salt and acid 

 he represented as follows : 



Metaphosphate of soda NaO.POs. 



Metaphosphoric acid - - H O . P 05. 



Speaking of the acid obtainable from, and by its neutralization recon- 

 verted into, the phosphate, pyrophosphate, and metaphosphate of soda 

 respectively, Mr. Graham remarked : "The acid, when separated from the 



