OF THE EXPEDITION. xiii 



" As they are of much weight, there are collateral reasons which it is proper that I should 

 state to you. On the second page of the preface to his Venus durchgang, Encke says: 'Until 

 1874, there is no hope of obtaining in this manner (by transits of Venus) a more accurate 

 knowledge of the solar parallax. The celebrated measures for the parallax of Mars, in the 

 middle of the last century, gave that of the sun within the eighth or twelfth part ; but this 

 latter method is of no other practical use than as it encourages a hope that, by a greater per- 

 fection of instruments, we may become more independent of the transits of Venus, which occur 

 so seldom.' And on page 109, he says : 



" ' The probable error in the observation of a contact being seven seconds, is so great as to ren- 

 der the determination of the sun's parallax within 0".01 almost hopeless for the next two centu- 

 ries. Such precision would have required a hundred observers at Wardhus, and a like number 

 at Otaheite ; whilst the difficulty of placing even thirty good and independent observers in the 

 vicinity of each of the best stations is so great, that it can hardly be expected we shall attain 

 such accuracy from any one transit. 



" ' All the observations of 1761, together, have but a value equal to three complete observa- 

 tions of the duration at Wardhus, compared with as many at Otaheite. If the weather had 

 been good at the eight northern stations in 1Y69, and there had been eight good observations 

 of the duration at the Friendly islands, these eight comparisons would have been as valuable as 

 the two hundred and fifty observations of the two transits actually made. In comparison with 

 that of 1769, the next two transits will be so unfavorable, that nothing short of perfection in 

 the construction of instruments, and art of observing, can be expected to compensate for it.' 



"We must, therefore, assent to the present calculated parallax for more than a century and a 

 half longer ; make another essay with Mars, or adopt the plan proposed by Dr. Gerling. Now, 

 does the present state of astronomical science require this trial by us, or will it be advanced 

 thereby? I feel confident that you, and all who reflect on these facts, will agree with me, that 

 to confirm the accepted 8".5776 within 0".09, even should no other astronomical result be derived 

 from it, is well worthy the expedition proposed to southern Chile, and the honor of its achieve- 

 ment not unworthy of American emulation. Moreover, though I need not have reminded you 

 of it, perhaps, should the observations be continued through the stationary terms of Venus in 

 December and January, 1850-'51, as several astronomers regard essential, the intermediate op- 

 position of Mars will occur, which may be made use of with greater probability of success than 

 at the Cape of Good Hope on the previous occasion, it being at that time the summer of the 

 southern hemisphere ; and we shall also be enabled to obtain a parallax in right ascension, as 

 well as in declination. 



" A few words more directly in reply to the two difficulties suggested ascertain to prove 

 obstacles in attaining ' results comparable in accuracy with the old determinations of the paral- 

 lax from the transit of Venus,' and I shall have done. And first, as to the probable error of 

 absolute declinations from a series of observations. Messrs. Bessel, Struve, Airy, and Arge- 

 lander, are such authorities, it would, apparently, be highly presumptuous to say that any one 

 could make better measures. But did either of those distinguished astronomers ever follow up 

 a star in declination for a year ? I greatly question it, as refers to the Germans, and know 

 (from the Greenwich observations) that Mr. Airy leaves such, indeed ninety-five hundredths of 

 all observations, to his assistants, with whose accuracy I trust I may be permitted to compare, 

 for, if what Mr. Walker says be true, my probable error is less than that of any known ob- 

 server. Enckc admits, page 30 of the volume quoted, that the probable error of observa- 

 tion at a station may be only 0".5. But I propose a form of instrument and method of observa- 

 tion which, united, will necessarily cause many of the ordinary discordances of observation to 

 disappear, and thus your apprehensions on this point be quieted ; that is. I want a circle whose 

 axis shall form a part of its telescope, and intend the nadir point to be examined immediately 

 before and after each series of observations. This construction of the telescope to me presents 

 one, or at most two, disadvantages. The first and most important is, greater instability in the 



