DISCUSSION OF ANALYSES. 79 



in its mineral content since the earlier analysis was made. The follow- 

 ing letter was received from the owner: 



There is no doubt in our opinion that mineral springs change in the different sea- 

 sons of the year in regard to the water and flow. We have also noticed changes in 

 the mineral waters or mineral springs in this section of the country before a storm or 

 heavy atmosphere. 



MARDELA WATER (No. 904 I. & W.). 



There is considerable variation in the two analyses of this water, 

 especially in the iron figure, the advertised analysis showing 138.1 

 parts per million and the Bureau of Chemistry only finding 13.6 parts. 

 It is of course possible that the spring, as it issues from the ground, 

 contains the quantity of iron claimed, but if such is the case it had 

 precipitated out before bottling the sample purchased by the Bureau 

 of Chemistry. 



VITAN TABLE WATER (No. 905 I. & W.). 



While there is considerable variation between the two analyses of 

 this water the mineral content is so small that the variation may be 

 considered as of no significance. 



BLUE LICK WATER (No. 1009 I. & W.). 



This sample is so highly mineralized that such difference as exists 

 between the advertised and Bureau of Chemistry analyses would prob- 

 ably be expected on account of the samples having been taken at differ- 

 ent times. Following is the owner's letter: 



In reference to the Blue Lick, which you recently analyzed, we have been in corre- 

 spondence with Messrs. & Co., from whom you purchased the water, and they 

 state that they have been in the mineral- water business now almost three years, and 

 that they found the water in the stock of waters they bought of their predecessor, 

 and according to our books this water was shipped from here in a wooden barrel 

 about three years since, and our water does not keep very well in wood for over six 

 months. The consequences are that the water you had to examine was of very poor 

 quality and certainly not up to the standard. 



SUBLETT LlTHIA WATER (No. 1010 I. & W.). 



While there is considerable variation between the advertised and 

 Bureau of Chemistry analyses, the difference could easily be accounted 

 for by changes in the composition of the water from time to time. 

 This appears as the most likely explanation in view of the owner's 

 statement in the following letter. It is a misnomer to call this a lithia 

 water. 



Professor came to the spring from Richmond and analyzed it. The capacity 

 of this spring is 1 gallon per minute in ordinary weather, but if we have exceedingly 

 wet weather it causes the spring to increase its flow. We never bottle any of the 



