THE VOYAGE OF H.M.S. CHALLENGER. 
130 
Remnant of water left —200 c.c., which would have barely sufficed for one alkalinity 
rmination. We preferred to utilise it for duplicate determinations of the lime and 
nil which gave, per 5 1 *3 1 gnus, of sea-water, 34*1 and 33’ 5, mean 33*8 mgrms. 
. .f < rudi lime, and 0'3332 and 0*3332 grm. of pyrophosphate of magnesia. The deposit 
•■■•ntained 0T247 grm. of lime, and 9*48 mgrms. of magnesia. The water under con- 
:■ it i< >n is one of the 77 which were completely analysed. (See table, page 23.) 
Assuming that all the 1215 c.c., which had been used already when the remnant of 
200 c.c. was analysed, had had the composition reported in the table referred to, 
we have — 
Per 100 parts of Chlorine. 
Remnant of 200 c.c. 
Part completely 
analysed. 
Original "Water. 
Average Surface 
Water. 
Lime, 
2-931 
3103 
3-499 
3-018 
Magnesia, 
11-442 
11-080 
11-163 
11-203 
The results for the quantities of lime are presumably less exact than those for the 
magnesia; and yet the former are easily interpreted, while the latter are difficult to 
understand. The water, it appears, was originally very rich in lime, and readily 
d'po-itc | j . i rt of this component as carbonate; but it is difficult to believe that the 
magn< 'i;t, after having been precipitated from part of the water, should have dissolved 
in the remainder to produce the large quantity of 1 1 *44 per 100 of chlorine. It is 
p mark il »1«- that in this case, as in that of No. 5, the lime, though a stronger base and 
I ip i»t in th<- u iter in a less proportion than the, magnesia, is preferably precipitated. 
l''r«>m our Table I. in this chapter it is seen that the alkalinity in our samples rarely 
- beyond GO mgrms. per litre; in one of the abnormal samples it comes up to 
75*88. What is the maximum value which the alkalinity might attain in the most 
liable < rcumstances ? This question uggested the following two experiments, which 
- ' !• w tli part of the supply of surface water which was referred to on page 115 
as having been collected for me near Ailsa Craig. 
/ • "t I. \ quantity of Ail- a Craig Water was saturated with carbonic acid at 
' 260 i c. then were mea ured off and digested in a stoppered bottle with two 
deal ••mat. "I lime for forty-eight hours at the ordinary temperature, 
agitation. Tie- -till turbid mixture was then filtered, and the alkalinity 
of 250 c.c. of the filtrate determined as usual. 
