452 Royal Society. 



best data for atomic weights. It is, however, pretty nearly compatible 

 with 23 or 24 atoms of water, according to Berze'ius's weight of the 

 atom of arsenic • and if the latter estimate be correct, it is curious 

 that the subarseniate differs from the neutral arseniate, merely by the 

 substitution of an atom of soda for an atom of water • for the latter 

 salt contains 25 atoms of water. 



The author's experiments on the composition of the subarseniate of 

 soda, — the results being reduced to the hypothesis, that it contains 

 23 atoms of water, — represent it as formed, (per cent.) of arsenic 

 acid, 27*69- soda, 2255 • water, 49*75. The subphosphate of soda 

 was found to consist (per cent.) of subsesquiphosphate, 4397 • water, 

 56*03 • the soda in the salt amounting to 24*87. 



The author attempted to determine the quantity of phosphoric acid 

 in this salt, by direct precipitation by nitrate of silver, but could not 

 obtain rigorously accurate results ; for the subphosphate of silver car- 

 ried down with it a portion of the nitrate, which washing could not 

 entirely separate. He likewise failed in his endeavours to obtain pure 

 subphosphate and subarseniate of potash. 



The subarseniate of barytes appeared, by a single experiment, to be 

 composed, (per cent.) of arsenic acid, 32*06 • barytes, 67 94 : from 

 which the salt would seem to contain an excess of base ; for by theory, 

 the subsesquiarseniate of barytes should be composed of, acid, 33*4 j 

 base, C6*6. 



When solution of muriate of lime is poured into an excess of solu- 

 tion of phosphate of soda, or when phosphate of lime, dissolved in 

 muriatic acid, is precipitated by ammonia, a gelatinous mass is formed, 

 which has been called the subphosphate of lime of bones -, the com- 

 position of which is singular, consisting, on the simplest view that can 

 be taken of it, of 3 atoms of phosphoric acid, and 4 of lime. It was 

 noticed by Berzelius. 



The author thinks the anomalous composition of this salt may in 

 some measure be explained by considering it as consisting of 1 atom 

 of the neutral, and 2 atoms of the subsesquiphosphate. According 

 to Berzelius, calcined ox-bones are composed of such a phosphate of 

 lime, with a little carbonate of lime • but a doubt arises of the accu- 

 racy of this view, from the fact, that the presence of carbonic acid in 

 the calcined phosphate of bones is no proof of the existence of that 

 acid in the same, previous to calcination. 



The earth of bones, alter calcination at a high temperature, con- 

 tains phosphoric, and not pyrophosphoric, acid ; the excess of base 

 preventing the transition. 



The author's analysis of subarseniate of lead, formed by the gradual 

 addition of solution of acetate of lead to solution of subarseniate of 

 soda, afforded a striking confirmation of the atomic weight of arsenic, 

 deduced by Berzelius from his analysis of arsenious acid by sulphur. 



A-paper was read, entitled, " Some Observations on the Structure 

 of Shells, and on the Economy of Molluscous Animals." By John 

 Edward Gray, Esq., F.R.S. 



The author distinguishes two kinds of structure in shells • the one 

 in which the calcareous matter is crystallized, composing what Mr. 

 Hatchett has called the porcellaneous structure • and the other, in 



