6 Thorpe and Simmonds, Lead Fritts. 



And with nearly equal percentages of silica : — 



9-- 37-6 37"9 28-0 



lO" 379 53"2 20 



Thus two fritts, A and B, may have the same quantity 

 of silica, yet A is largely attacked and B but slightly 

 (Nos. 9 and 10). Or A may have more silica than B, and 

 still be unprotected (Nos. 5 and 6 ; also 7 and 8). Finally 

 A may have less silica than B, and nevertheless be the 

 more largely attacked (Nos. 5 and 10). It is difficult to 

 give much credence to a theory of protective action which 

 is certainly not exercised at all in many cases, and in 

 others exhibits so much caprice as to render impossible a 

 reliable forecast of its probable effect. 



Fritt not a single compound. — Underlying the whole 

 of Messrs. Jackson and Rich's argument is the assumption 

 that a fritt is a single chemical entity. There is evidence 

 to show that this is probably an erroneous assumption. A 

 fritt would appear, in fact, to be usually a mixture of at least 

 two lead compounds. The proportion of one of these, how- 

 ever, may be small compared with the main bulk of the fritt. 



The considerations which suggest this are as follows : — 



I. Chemical. — In so far as a fritt is attacked by a 

 solvent, its soluble constituents, if the fritt is a single com- 

 pound, will be found to bear the same proportion to one 

 another in the solution as in the original fritt. Whether 

 the silicic acid be dissolved or not, this relation will hold 

 for the base-oxides. If, however, the bases in the 

 dissolved portion be found to have a different proportion 

 from one another than exists in the original fritt, then the 

 result is evidence that the fritt was not a single compound. 

 Applying this deduction to the case of seven specimens 

 ot fritts whose composition had been ascertained, together 

 with that of the dissolved portion given on treating the 

 fritts with excess of dilute hydrochloric acid, the following 

 comparisons were obtained : — 



