128 
THE VOYAGE OF H.M.S. CHALLENGER. 
Notes on the Anomalous Cases. 
T unbr.i . keletl Nos. are tlio Challenger numbers; those in square brackets the laboratory numbers 
of the respective waters). 
No. 5 [204], page 125. A surface water. The alkalinity determination in two 
titrations led to the low values of 28 ’GO and 28 'GG per litre. The inside of the bottle 
was coated with a crystalline deposit up to a line corresponding to 1392 cubic 
icutimctres. Actual contents = 802 c.c. of sea-water, of which 500 c.c. served for the 
two analyses referred to. 'Flic deposit, after having been rinsed twice with distilled water, 
. is dissolved in hydrochloric acid (which caused an evolution of carbonic acid), and the 
lime and magnesia in the solution determined successively by means of oxalate of 
ammonia and phosphate of ammonia respectively. The lime (CaO) amounted to 0*1173, 
(MgO) to 0*01964 grin. According to table, page 43, x=20'461. The 
lime and magnesia in the water wov determined by the method used in the 77 complete 
analyses (see pp. 9, &c.), and, in 51 '26 grms. of water, found to be as follows: — 
Crude lime = 33‘l and 337, mean 33'4 ; magnesia, as pyrophosphate = 0 ’33 01 and 
0*3308, mean = 0*3304. Now, correcting the lime by multiplication with 0'91, and 
assuming that the results for it and the magnesia hold for all the 1392 c.c. of water 
Daily in the bottle, we arrive at the results, given in the following table, and con- 
t • t • d therein with the mean values deduced from the 77 complete analyses, as stated 
at the end of this chapter. 
Present in Grms. 'per 100 Grms. of Chlorine in — 
Deposit. 
Water. 
Total. 
Mean of 77 
Sea-waters. 
Lime, CaO, 
0-401 
2-899 
3-300 
3-026 
Magnesia, MgO, . 
‘ 1 
0-0671 
11-352 
11-419 
11-212 
The numbers under “ Water ’’ are probably a little too low, because the part of the 
'' r wh ch had been taken out of the bottle before the present analyses were made 
must be presumed to have been richer in lime and magnesia than the remaining 802 c.c. 
But even with the numbers as they stand, this water would appear to have been 
ri- li in both components originally. Supposing the deposit were rcdissolved 
in 1392 C.C. «-f water like the remnant analysed, the alkalinity per 100 of chlorine 
u • • * 1 1 ‘ f l 1 ■ by 0 389, or by 0*2 1 5G per 100 of salts, which, when taken together 
