132 RE-EXAMINATION OF AMERICAN MINERALS. 



was found to be an important ingredient in its constitution ; 

 in his analysis, however, a large amount of alkalies is indi- 

 cated. As the results of our analyses show but a trace of these 

 substances and no other base but lime, it is possible that Dr. 

 Erni's alkali determination was made through mistake on some 

 of the feldspar accompanying danburite, as it not unfrequently 

 happens that the granular portions of the feldspar resemble the 

 lighter varieties of danburite. This supposition appears rea- 

 sonable, as the silica and lime of his determinations agree with 

 those about to be stated. The composition as given by the 

 authorities mentioned are 



Si Ca £l 

 56.00 28.33 1.70 Y?0.85 K (with Na?) and loss 5.12 H8.0 =100. Shepard. 

 49.74 22.80 Mgl.98 £e and Si 2.11 K 4 31 Na 9.82 B 9.24=100. Erni. 



The results of the analyses just made are as follows : 



12 3 4 



Silica 48.10 48.20 



Alumina , 



Peroxide of iron... \ 'r 1-02 

 Manganese ... 



Lime 22.41 22.33 22.22 22.11 



Magnesia .40 undeterm. 



Boracicacid 27.73 27.15 



} - 80 } 



.56 J 



Ignition .50 .50 



100.00 99.20 



This corresponds to the following constitution : 



Per cent. Oxygen ratio. 



4 atoms silica 49.42 4 



3 •• boracicacid 28.02 3 



3 •• lime 22.56 1 



It is not easy to decide upon the formula by which danburite 

 is to be expressed. The same difficulty occurs here as in the 

 case of datholite, Berzelius considering the lime only as acting 

 the part of base, while Eammelsberg regards the boracic acid 

 also as performing the part of a base. The formula for dan- 

 burite under these views would be expressed by 

 2CaSi 2 +CaB 3 or Ca 3 Si 2 +B 3 Si 2 . 



We are in favor of regarding the latter as representing the 

 true formula. 



If danburite be examined in its relation to datholite, it will 

 be found to differ from the latter in having just one half the 

 number of atoms of lime minus three atoms of water. 



