INTRODUCTION. vii 



The observations and reductions, I found, had been printed in great detail, 

 especially for the district of the Sandwich Islands. It was desirable that this 

 should be done to some extent as a general specimen of the operations ; but 

 I now determined to print the remainder on the scale which I had intimated 

 in an address to the Royal Astronomical Society, published in their Monthly 

 Notice for 1875, March 12, " In the accounts of transits it is sufficient to 

 " give description of instruments and methods, constants of adjustments, 

 " and tables of clock-errors, and analogous abstracts of comparisons of 

 " chronometers, &c. But it is necessary to give in the fullest detail 

 " everything that bears upon the actual observation of contacts, or upon 

 " the observer's impression at the time of making the observation, or upon 

 " the micrometer-measures, or upon the photographs and the measures 

 " of the photographs, &c. ; with sufficient description of the instruments and 

 " their adjustments at the time. Clock Time and Local Sidereal Time are to 

 " be given for every observation." 



The carrying out of this change, though in its main feature it is a very 

 large diminution of the matter prepared for 'press, yet, as it consisted of 

 sometimes extracting numbers, sometimes taking the means of numbers, 

 always requiring numerous references to the originals and to the first 

 calculations founded on them, has in reality employed much of my time. 

 It has, indeed, occupied all the hours, not engaged on routine business, on 

 which I could usually have reckoned for other matters of science. 



In regard to the form in which the Results are to be presented to the 

 reader, I have thought it best to leave the reduced observations in the state 

 in which he will find them near the end of each Part. I have endeavoured to 

 give the Equations in the shape which will admit of combination in the 

 easiest way for the computer's further operations (whether he may desire to 

 use the Calculus of Probabilities for the whole, or to make any special 

 selection of combinations) when he shall have decided on the recorded 

 phase of contact of limbs which he thinks best to adopt. The numerical 

 value of the first term of each Equation (on which all depends) will be 

 adapted to any time differing from that which I have used, by merely 

 expressing numerically in seconds that difference of time, and substituting it 

 for the symbol 8 t. 



In regard to the photographs, of which I have given, in the Appendix to 

 this work, a general account, with sufficient details on the instruments 



