KEW OBSERVATORY AND METEOROLOGY 227
perfectly, to a considerable improvement in the make of the cheaper sextants.
Another thing that I did was to contrive an apparatus by which thermometers could be rapidly and yet very accurately verified, and by which from ten to twenty thousand clinical thermometers are still annually tested. Mr. De la Rue gave me help in devising this. The few pence gained on each of these many thermometers amounted to a respectable sum, and confirmed the solvency of the institution, whose margin of profit over loss was always small and had been precarious. We were thus in a better position to extend our work and to add to our instruments, and we did so.
Another operation which I was among the first, if not the first, to suggest, was the rating of watches. This has been a real success. The performances of watches, when we first took the matter in hand, was by no means proportionate to their cost, more than one highly ornamented and expensive time-keeper failing to obtain a class-place equal to that of others of much inferior pretensions. Now a Kew certificated watch has a special and recognised value, and the makers of valuable watches are far more on their mettle than they used to be.
The influence of the Kew verifications as time went on extended in many other directions, as by testing the performance of telescopes and opera-glasses supplied to the army and navy, in order to ascertain whether their capabilities were up to the specified standard. Mariners' compasses of complicated and delicate construction were also dealt with. A beautiful apparatus devised by Sir Wm. Abney and Major