OP ARTS AND SCIENCES. 73 



tungstates and metamolybdates. On the other hand, in by far the 

 greater number of compounds thus far studied, the complex is to be 

 regarded, I think, as built up of terms of the form ROo, which may 

 or may not be linked together by one or more atoms of oxygen. In 

 certain cases, therefore, vanadic pentoxide may enter the complex 

 with the structural formula 



1 



I 



o 

 i 

 — V = 0., 



I I/O 



1 )v — O— V^ I 



This last view is strongly supported by the frequent presence of 

 vanadic dioxide, two molecules of which may be considered as associ- 

 ated without a link of oxygen, as in the structural formulas 



I 



— VO., O I I o 



I and I ;V — V:; I . 



— YO, 0^1 1^0 



Finally, as I shall also prove, the structural formula must show 

 what particular modification of phosphoric or vanadic oxide or acid is 

 contained in the salt, since we have not merely ortho-phospho-tung- 

 states but pyro-phospho-tungstates, various modifications of meta- 

 I^hospho-tungstates, and corresponding compounds of molybdenum. 

 It is at least probable that the different modifications of vanadic pent- 

 oxide or acid enter in a similar manner. These appear to be very 

 numerous, and a wide field is thus opened for research. I shall return 

 to this subject in the purely theoretical portion of my work. 



If the view which I have taken as to the composition of the com- 

 plex acids containing vanadium be correct, we ought to find some 

 additional support for it in the constitution of the salts of vanadic 

 acid. These have been studied by Roscoe, and especially by Nor- 

 blad * and by Rammelsberg,t but the results obtained b}' these chem- 



* Bidrag till kannedomeii om Vanadiums Amfidsalter. Upsala Universitats 

 Arsskrift, 1873. 



t Sitzungsbericlite der KGnigl. Preuss. Akaderaie der Wissenschaften, 1883, 

 p. 1. 



