230 
THE VOYAGE OF H.M.S. CHALLENGER. 
mi nations are of little use henceforth, unless carried out with a multitude of freshly- 
drawn samples, and coupled with alkalinity determinations. The difficulty is to discover 
;i nn thod which would combine high precision with sufficient ease and rapidity of execution. 
My method, described in the memoir as the “ vacuum method,” would work as easily 
as Buchanan's did on board ship, but either is troublesome, and would become very 
tedious if duplicate or triplicate analyses were demanded, as they ought to be. The 
most practical plan, perhaps, would be to combine the determination of the carbonic acid 
with that of the nitrogen and oxygen, as proposed by me on page 105 ; that is to boil 
out the gases in Jacobsen’s apparatus in the presence of hydrochloric acid, to seal them 
up, and subsequently analyse them at home. 
4. The Absorbed Oxygen and Nitrogen. — Jacobsen’s method is the only one which 
would work on board ship, and it certainly is susceptible of a fair degree of exactitude. 
But in any future expedition it would be desirable to have all gas-extractions done in 
duplicate or triplicate, in order to supply the one item without which no series of analyses 
can be properly discussed, namely, the probable error in the single determination. 
Meanwhile the best thing that could be done in regard to all the analytical problems 
referred to would be to work many times on samples of the same kind of water, with a 
view of improving upon the methods and ascertaining the extent to which that one 
water fluctuates in its composition. 
