70 PROCEEDINGS OF THE AMERICAN ACADEMY 



cobalt, or nickel, and of sodic di vanadate, dark green crystalline pre- 

 cipitates are formed on standing, which would appear to be vanadico- 

 vanadates, but which, as tliey are formed in the presence of an excess 

 of a strong acid, may prov^ to be types of vanadico-vanadic acids. In 

 his valuable paper on the salts formed by vanadic acid, Rammelsberg 

 has described a remarkable compound, which may be regarded as an 

 anhydride of a special form of vanadico-vanadic acid. This compound 

 has the formula 



VA + 2V0,, 



and forms black microscopic crystals. Its structural formula may be 

 written : 



It is easy to see how acids of various types might be derived from 

 such anhydrides by replacing oxygen by an equivalent of hydroxyl. 

 The existence of very well defined series of phosphoroso-tungstates and 

 phosphoroso-molybdates, and of similar series of compounds containing 

 AsgO., and SbgOo , would naturally lead to the attempt to form similar 

 classes of complex acids containing VoO„. Want of material has pre- 

 vented me from experimenting in this direction, and I must content 

 myself with directing the attention of other cliemists to the subject. 

 Vanadic protoxide appears to possess the properties of a base only, but 

 our knowledge of the subject is so imperfect that it may also be well 

 worth while to determine whether there may not be a class of vanadium 

 compounds corresponding to the hypophospho-tungstates and hypo- 

 phospho-molybdates. 



The existence of classes of phospho-vanadates and ai'senio-vanadates 

 embraced under the general formula 



EoO- . V2O5 . n 11,0 



would seem to imply that of corresponding chlorides, bromides, etc., 

 of corresponding or analogous types, 



R'Cls . RCl, , R'Cl, . RCl, , etc. 



Various compounds of this kind have already been observed by chem- 

 ists, though, so far as I am aware, no attempt has been made to classify 



