He limited his task to fixing the true period of the sun's rotation (of which curiously discrepant values had been obtained), to tracing the laws of distribution of maculæ, and investigating the existence of permanent surface-currents. Adequately to compass these ends, new devices of observation, reduction, and comparison were required. Leaving photography to his successors as too undeveloped for immediate use, he chose a method founded on the idea of making the solar disc its own circular micrometer. An image of the sun was thrown upon a screen placed at such a distance from the eyepiece of the 4½-inch equatoreal as to give to the disc a diameter of 12 to 14 inches. In the focus of the telescope, which was firmly clamped, two bars of flattened gold wire were fastened at right angles to each other, and inclined about 45° on either side of the meridian. Then, as the inverted image traversed the screen, the instants of contact with the wires of the sun's limbs and of the spot-nucleus to be measured were severally noted, when an easy calculation gave its heliocentric position (ib. xiv. 153).
In this manner, during seven and a half years, 5,290 observations were made of 954 separate groups, many of which were besides accurately depicted in drawings. By the sudden death of his father, however, in July 1858, and the consequent devolution upon Carrington of the management of the brewery, the complete execution of his project of research was frustrated. He continued for some time to supervise the solar work he had previously carried on in person; but in March 1861, seeing no prospect of release from commercial engagements, he thought it advisable to close the series. The results appeared in a 4to volume, the publication of which was aided by a grant from the Royal Society. Its title ran as follows: ‘Observations of the Spots on the Sun from November 9, 1853, to March 24, 1861, made at Redhill’ (London, 1863). Never were data more opportunely furnished. Perhaps more effectually than the pronouncements of spectrum analysis, they served to revolutionise ideas on solar physics.
Efforts to ascertain the true rate of solar rotation had been continually baffled by what were called the ‘proper motions’ of the spots serving as indexes to it. Carrington showed that these were in reality due to a great ‘bodily drift’ of the photosphere, diminishing apparently from the equator to the poles (ib. xix. 81). There was, then, no single period ascertainable through observations of the solar surface. By equatorial spots the circuit was found to be performed in about two and a half days less than by spots at the (ordinarily) extreme north and south limits of 45°. The assumed ‘mean period’ of 25.38 solar days applied, in fact, only to two zones 14° from the equator; nearer to it the time of rotation was shorter, further from it longer, than the average. Carrington succeeded in representing the daily movement of a spot in any heliographical latitude l, by the empirical expression 865′ ± 165 . sin 7/4 (l – 1°). But he attempted no explanation of the phenomenon. It formed, however, the basis of Faye's theory (1865) of the sun as a gaseous body ploughed through by vertical currents, which finally superseded Herschel's idea of a flame-enveloped, but cool, dark, and even habitable globe.
Carrington's determinations of the elements of the sun's rotation are still of standard authority. The inclination of the solar equator to the plane of the ecliptic he fixed at 7° 15′; the longitude of the ascending node at 73° 40′ (both for 1850) . A curious peculiarity in the distribution of sun-spots detected by him about the time of the minimum of 1856, afforded, as he said, ‘an instructive instance of the regular irregularity and the irregular regularity’ characterising solar phenomena (ib. xix. 1). As the minimum approached, the belts of disturbance gradually contracted towards and died out near the equator; shortly after which two fresh series broke out, as if by a completely new impulse, in comparatively high latitudes, and spread equatorially. No satisfactory rationale of this curious procedure has yet been arrived at. It is, nevertheless, intimately related to the course of sun-spot development, since Wolf found evidence of a similar behaviour in Böhm's observations of 1833–6, and it was perceived by Spörer and Secchi to recur in 1867.
While still in his apprenticeship at Durham, Carrington repaired to Sweden on the occasion of the total solar eclipse of 28 July 1851, and made at Lilla Edet, on the Göta river, observations printed in the Royal Astronomical Society's ‘Memoirs’ (xxi. 58). The experience thus gained was turned to public account in the compilation of ‘Information and Suggestions addressed to Persons who may be able to place themselves within the Shadow of the Total Eclipse of the Sun on September 7, 1858,’ a brochure printed and circulated by the lords of the admiralty in May 1858. The eclipse to which it referred was visible in South America. Besides his friend, Mr. Hodgson, he was the sole witness of the extraordinary solar outburst of 1 Sept. 1859. His account of an observation memorable in the history of solar
