530 THE MEDICINAL SPRINGS OF SOUTH AFRICA. 



data necessary for making such a distinction not being available 

 or existent, the course adopted was to regard the combined car- 

 bon dioxide as hydrocarbonate ion only in all cases where 

 calcium, magnesium, and ferrous iron are present. In the case 

 of the alkaline waters, however, whose predominating con- 

 stituent is sodium carbonate, the combined carbon dioxide was 

 calculated as carbonate ion. 



In many of the analyses under review the values of the 

 alkalies were not determined. After satisfying all the bases, the 

 excess of acids was assigned to the oxides of the alkalies, which 

 are put into reckoning as NaaO. Jn such cases, and generally 

 s])eaking in all cases where there was reason to believe that the 

 values of some of the constituents had been obtained by calcu- 

 lation instead of by actual determination, there could be no ol)ject 

 in recalculatinji: the values in terms of ion-equivalents. 



Where such tables of ion-equivalents have been made they 

 are appended to the descriptions of the individual springs in Part 

 II of the ])aper. The values are obtained by dividing the weight 

 of each ion (expressed in grams per 100,000) by its atomic 

 weight (or the sum of the atomic weights), and multinlying by 

 the valency in order to obtain monovalent ions. 



In accordance with the generally accepted theory, alkali 

 sulphides are regarded in my calculations as being ionised into 

 scdium (or potassium) on the one hand, and inti) monovalent 

 SH ions on the other. 



Ga.yc.s-. — The o])eration of collecting the gases escaping from 

 a spring is a difficult one, and cann'>t be entrusted to a layman. 

 Dissolved gases diffuse out in the course of a few liours. As in 

 most cases the waters were not collected l)y the analysts them- 

 selves, and were transported a considerable distance before being 

 examined, our information about the gaseous constituents is most 

 incimplete. In the case of sulphur springs the values for sul- 

 Dhuretted h\drGgen are often not given at all in the original 

 analvses. If recorded, they should be regarded with a certain 

 .'iniMunt I if reserve, unless the estimptinn was cari'ied oiU at the 

 .•-itc of the spring. 



RadioactiTity. — It is impossible *.o ex])ress an opinion on 

 this pnmt and to make comparisons, unless (piantitative data 

 aljout tlie radioactivity are available. ex])ressed in terms of an 

 accejjted international standard. 



Medicinal Properties. — In the al)sence of delinite ex|)ressions 

 of opinion from the medical ])rofession, my information is based 

 (;n the statements of guide-books of South Africa, on the evi- 

 dence of patients, and on current beliefs. To what extent such 

 infi^-rmation is credible can be gathered from the fact that prac- 

 tically all our si)rings are su])]>osed to be s])ecifics for rheumatism 

 and ciuaneous diseases, no matter wXvAt their com])<)sition. 



CoDiparison ivitli linropcan Springs. — The original rejiorts 

 in some cases ])oint out the similarity of the waters examined to 

 well-known F,ur()])ean medicinal springs. These statements have 

 liecn includerl in the descrijnions of the individual springs in 



