the absolute instant of any nominal mean time computed in the 

 beginning of 1883, and that this error accumulates at the rate of 

 l"48 s per annum. It was in vain that Airy, Adams, and Cayley 

 endeavoured privately to convince Stone of the error of this con- 

 clusion, and to show him that the effect of the change in question 

 was l/365th part of what he claimed it to be. Adams then pub- 

 lished a paper,* which explains the whole point at issue in the 

 clearest and simplest manner, viz., that mean solar time is measured, 

 not by the sun's mean motion in longitude (as Stone's theory sup- 

 poses), but by the motion of the mean sun in hour angle, which is 

 about 365 times greater in amount. Cayley, Gaillot, and afterwards 

 Newcomb, published notes and questions intended to put the matter 

 in a way that should convince Stone of his error. But no arguments 

 or illustrations were of avail, and, during the last ten years of his 

 life, almost the whole of his time which was not occupied by pressing 

 official duty was employed in devising fresh methods of presenting 

 his theory for discussion, or in comparing the places of the moon, as 

 observed at Oxford, with Hansen's tables both corrected and uncor- 

 rected for the term dS given by Stone's theory. By a very remark- 

 able coincidence the present errors of Hansen's Tables of the Moon 

 agree within very narrow limits with the corrections dS, and it must 

 have been this circumstance which completely blinded Stone's eyes 

 to the true state of the case. What Stone's theory does not explain 

 is the simple fact that if no change had been made in the ' Nautical 

 Almanac ' in 1864, or, what is the same thing, if the mean solar 

 times derived from Carlini's (or Bessel's) tables had been used to 

 convert the sidereal times of observed transits of the moon at Oxford 

 into Greenwich mean time, and if the mean times so computed had 

 been used as arguments to derive the tabular place of the moon from 

 Hansen's tables, it would be found that the errors of Hansen's tables 

 in right ascension would have differed but a few hundredths of a 

 second of time from what they would have been if computed from 

 the data of the present ' Nautical Almanac' 



On taking up residence at Oxford, Stone found that the staff con- 

 sisted of but one assistant, whose time was chiefly occupied in making 

 and reducing the regular meteorological observations and observing 

 stars for clock error. Two assistants were added to the staff, a 

 systematic examination of the transit circle was instituted, and 

 regular meridian observing was begun in 1880. Plans were made 

 for the construction of a catalogue of all stars to the 7th magnitude 

 from N.P.D. 115° (the northern limit of his Cape zones) to the 

 equator. Observations were secured at a rate varying from about 

 2000 to 3500 per annum from 1881 to 1893, these were regularly 

 reduced and published in annual volumes, and in 1893 meridian 

 * ( Monthly Notices R.A.S.,' vol. 44, p. 43. 



