164 The Philippine Journal of Science i9i6 



are calculated separately for different minutes. For instance, 

 during the first minute the iron is only 0.64 per cent of the 

 total metal dissolved. During the fifth it is 20.29 per cent; 

 in the seventh, 84.86 per cent ; in the eighth, 91.32 per cent. This 

 work seems to indicate that, in addition to alloys like FeZn, 

 and FeZn^, there may be a series high in iron and low in zinc, 

 such as Fe^Zn and Fe„Zn. 



On the zinc curve, from Pq to P^, the acid is dissolving the outer 

 coating of almost pure zinc and the curve is nearly a straight 

 line. From P^ to P2 the velocity falls off rapidly because the zinc 

 coat has almost disappeared and the alloys are exposed. On the 

 iron curve the maximum rise is between P\ and P\, or after the 

 zinc is almost completely dissolved. 



As has been stated, the iron dissolved during the first few 

 minutes is due to the iron contamination of the molten zinc 

 before the sheet is placed in the bath. It is true also that during 

 that time the acid may have gone entirely through the outer 

 layer of zinc at some point and started acting on one of the 

 zinc-iron alloys. Therefore, if we know the amount of iron dis- 

 solved during the first stage of the reaction, or up to the point 

 of the break in the zinc curve, P^, we have a hint as to the 

 value of the sheet to withstand corrosion. A high iron content 

 would show that the iron contamination of the spelter is consid- 

 erable, or that the zinc coating is thin in spots. Either con- 

 dition would indicate that the sheet is of inferior quality. It 

 is obvious that the outer zinc coating should be as uniform as 

 possible, and that the purer it is, the better. On the other hand, 

 a very small amount of iron removed during the first stage of 

 the reaction would assure both the purity and the uniformity 

 of the outer coating. I propose to investigate this question 

 further and to attempt to determine the maximum percentage 

 of iron that is allowable in a good quality of galvanized sheet 

 under these conditions. Such a test would be simple to carry 

 out, but important. 



CONCLUSIONS 



1. A hydrochloric acid solution gives satisfactory results for 

 determining the amount of the coating on galvanized iron. 

 Directions for its use are given. These include the concentra- 

 tion of the acid, the temperature, and the size of the test piece. 



2. The accuracy and reliability of the test have been proved 

 in a number of ways. It requires little labor and is rapid and 

 inexpensive. 



3. A convenient method is given for correcting for the alloyed 



