XV1il Obituary Notices of Fellows deceased. 
approval, and it is noteworthy for the opinion expressed therein, that “ the 
large amount of information acquired in the laboratories of our great 
manufacturing concerns might well be published without any injury to the 
individual manufacturer.” Eighteen years later, when Warington had for a 
second time gone to work in Lawes’ tartaric and citric acid factory, he 
published another paper dealing with these acids, and with the detection of 
the presence of lead in them. With this solitary exception, all Warington’s 
subsequent work was on agricultural chemistry, and all of it was done in the 
Rothamsted laboratory. 
While still at Millwall, he had been writing a good deal on agricultural 
subjects—several articles for “ Watts’ Dictionary” and for the Agricultural 
and Horticultural Co-operation Association—and he had, moreover, as already 
mentioned, been in continual consultation with Lawes as to the Rothamsted 
results ; he was naturally, therefore, prepared to receive Lawes’ suggestion 
that he should go and work in the Rothamsted laboratory. The terms were 
all settled, and had readily been assented to by Warington; for, although 
they had involved a reduction of salary to two-thirds of that which he had 
been receiving at Millwall, he obtained a certain amount of freedom by way 
of compensation. He was to be at liberty to publish his own work in his 
own name, provided that it made its appearance as Rothamsted work ; but in 
cases where the work dealt with subjects which had already occupied the 
Rothamsted investigators, it was to be published in the joint names of Lawes, 
Gilbert and Warington. This arrangement, however, owing to some unfore- 
seen difficulties, was not carried out; and it was not till after a delay of two 
years that Warington went to Rothamsted (in 1876), under an agreement 
for a year only, to work simply as Lawes’ private assistant. The engagement 
was subsequently extended, and all his results were published, either in his 
own name or in the names of Lawes, Gilbert and Warington. 
Before removing to Harpenden, he went to work at the laboratory at South 
Kensington in order to learn water and gas analysis under Frankiand’s 
assistant, some of the Rothamsted soils being sent to him for practising 
determinations of nitrogen. While there he devised a method of extracting 
soils by the vacuum pump, which method has since been largely used at 
Rothamsted. In the autumn of the same year (1876) he made a short tour 
among the German experimental stations, and then took up his residence for 
good at Harpenden. 
The construction of a gas analysis apparatus (under Frankland’s direction) 
for the Rothamsted laboratory occupied a considerable time, and, pending its 
completion, Warington made a study of the indigo method of determining 
nitric acid. This method, as generally used, he found to be full of sources of 
error. The principal of these he succeeded in correcting, and, with the 
method of determination, thus rendered trustworthy, he proceeded to determine 
regularly the nitrates in the drainage-water from the various wheat plots in 
Broadbalk field. The chlorides were determined at the same time. No such 
systematic work had been previously done; whilst the methods of sampling, 
