September i, 1892] 



NATURE 



425 



mental means for the determination of the positions and motions 

 of the bodies included in the solar system. Accuracy in astro- 

 nomical methods and results did not exist, even approximately, 

 until after the revival of practical astronomy in Europe about 

 the beginning of the sixteenth century ; and, before the end of 

 that period, the crude instruments of the early astronomers 

 reached their highest perfection in the hands of the skilful genius 

 of Uraniborg. 



The invention of the telescope, the application of the pen- 

 dulum to clocks, the invention of the micrometer, the combina- 

 tion of the telescope with the divided arc of a circle, the 

 invention of the transit circle by Roemer, with many improve- 

 ments in minor apparatus, distinctly stamp the seventeenth 

 century as a remarkable period of preparation for the 

 achievements of the next century. 



From the standpoint of the modern mechanician the instru- 

 ments at the Greenwich Observatory in Bradley's time were 

 very imperfect in design and construction, and yet on the 

 observations obtained by his skill and perseverance depends the 

 whole structure of modern fundamental astronomy. The use of 

 the quadrant reached its highest excellence under Bradley's 

 management. 



The next advance, the real work with divided circles, began 

 at Greenwich in l8ll, under the direction of Pond. Since that 

 epoch, theory and observation have held a nearly even course in 

 the friendly race toward that elusive goal perfection ; and the 

 end is not yet. A careful, but independent, determination of 

 the relative right ascensions of the principal stars, supplemented 

 by a rigorous adjustment of such positions with regard to the 

 equinoctial points, and a similar determination of the relative 

 zenith or polar distance of the same bodies, finally referred and 

 adjusted to the equator or the pole, seem in this brief statement 

 to be, at least, simple problems. If, however, we examine the 

 conditions in detail the simplicity may not appear so evident ; 

 and this characteristic may prove to be one reason why this 

 important branch of astronomical research is now so generally 

 neglected. 



In the first place, it must be understood that such an investi- 

 gation cannot be completed in a few months. At least two and 

 preferably three years' work in observing are necessary to secure 

 ^ood results. Skilled observers, and not more than two with 

 the same instrument, are absolutely necessary. Such work can- 

 not be confided to students or beginners in the art of observing, 

 or to observers who have acquired the habit of anticipating the 

 transit of a star. The telescope and the circles, the objective 

 and the micrometer, the clock and the level must be of the best 

 quality, for imper.fections in any of these essentials render the 

 best results impossible. A thoroughly good astronomical clock 

 is the rarest instrument in the astronomer's collection. It is not 

 sufficient that a clock should have a uniform daily rate, the rate 

 should be uniform for any number of minor periods during the 

 'twenty-four hours. The absolute personal error in observing 

 transits should be determined at least twice a week, and when 

 it is not well established it should be found every day. The 

 level error should be found every two hours, and the greatest 

 care should be exercised in handling this important instrument. 

 The division marks should not be etched on the level tube 

 unless the values of the divisions are frequently examined, for, 

 sooner or later, such tubes become deformed on account of the 

 broken surface, and are then worthless. 



In the determination of zenith distances the effect of refraction 

 plays such an important part that no work can rightly claim to 

 be fundamental until the local refraction has been carefully in- 

 vestigated, and special corrections to the standard tables, if 

 necessary, have been deduced for each observing station. The 

 ordinary mode of observing temperature is quite inadequate to 

 the importance of the phenomena. These observations should 

 be made as near as possible in the mass of air through which the 

 objective of the telescope is moved, and also in the opening in 

 the roof and the sides of the observing room where the outside 

 air comes in contact with that in the building. The thermometers 

 should all be mounted, so that they may be whirled in that 

 portion of the air where the temperature is desired, and they 

 should be tested at least once a year to determine the change in 

 the position of the zero of the scale. But a complete list of the 

 things to be done, and of the errors to be avoided, are too volu- 

 minous for this occasion, and are not necessary to show the 

 complex character of the problem ; the suggestions already made 

 must suffice. 



For many years an immense number of observations of the 



NO. I 192, VOL. 46] 



larger or the so-called standard stars have been made at the 

 principal observatories, for diffisrent purposes and with varying 

 degrees of accuracy, but it is not certain that the work of the last 

 thirty years, with all the advantages of improved apparatus, has 

 resulted in more exact determinations of even the relative right 

 ascension of such stars. There can be no doubt that the chrono- 

 graphic registration of star transits has given more accurate 

 results for the smaller stars, but I think it is equally tru-; that, 

 in the case of first and second magnitude stars at least, no 

 improvement has been made in accuracy. 



With double threads it is possible to observe the zenith dis- 

 tances of such stars with a fair degree of precision, because the 

 operation is one of comparative deliberation, and the centre of 

 the mass of light can be placed midway between the threads 

 with little difficulty. But the attempt to note, with a chrono- 

 graph key, the instant when a swiftly-moving and irregular mass 

 of light, like a Canis Majoris or o Lyrae, is bisected by a transit- 

 thread, is an operation that rises but little above the level of 

 ordinary guesswork. Transits of first and second magnitude 

 stars cannot be observed with an objective of more than four 

 inches aperture with the desired accuracy, unless the apparent 

 magnitude is reduced, by means of screens, to that of a fourth or 

 fifth magnitude star. It is necessary in this connection to avoid 

 confounding the methods employed in the observations of the 

 bodies of the solar system with those for obtaining fundamental 

 places of the stars. The observations of the Sun, Moon, Mercury, 

 and Venus with a transit circle are, from the unavoidable con- 

 ditions, necessarily uncertain to a degree even beyond the 

 probable error involved in the observations of the large stars. In 

 spite of these unfavourable conditions, however, the continued 

 observations of these bodies at the principal observatories for 

 many years have produced the most valuable results, even when 

 the work on the standard stars, on which their results depend, 

 has no claim whatever to a fundair.ental character. 



In geographic exi)loration the first endeavour is to secure ap- 

 proximate positions of salient points from a rapid reconnois- 

 sance. This is followed by more careful work, fixing the 

 observing stations with that degree of precision which ensures 

 good results. Finally, the highest qualities of skill and science 

 are combined to exhaust all available means to reach the greatest 

 attainable accuracy. In the exploration of the heavens, the 

 first two of these steps have already been taken, and most of the 

 stars of the larger magnitudes have been so well observed, that 

 the accuracy of their positions is not only far higher than is 

 required by the greatest skill of the navigator, but it is equal to 

 all the demands of ordinary practical work. It is the next step 

 which challenges the skill of the mechanician, the observer, and 

 the computor ; and astronomers cannot rest at ease until all 

 known resources have been exhausted in the attempt to reach 

 the best results. It is not a very difficult matter to fix the posi- 

 tion of stars within a range, in the individual observations, of 

 three or four seconds of arc ; but that degree of accuracy is not 

 sufficient for the more exact problems of astronomy, and it falls 

 far short of what is required in the important discussions of solar 

 and stellar motions. 



Bradley's observations furnish the data for Bessel's " Funda- 

 menta Astronomise,"and many astronomers have since attempted 

 by reductions to obtain improved positions for Bradley's stars. 

 The value of these observations in the development of modern 

 astronomy can hardly be exaggerated. Their importance in the 

 dermination of stellar proper motions increases with the lapse of 

 time, and yet the accuracy of the original observations was far 

 inferior to that obtained in ordinary routine work with modern 

 methods and improved instruments. 



Fundamental catalogues of stars have notably increased since 

 the " Fundamenta Astronomise," but the demand has not yet 

 been satisfied. The catalogues of declinations or north-polar 

 distances are more numerous than those of right ascension, 

 evidently because, for many reasons, independent declinations 

 are more readily determined. 



There is probably no collection of the right ascension of the 

 large stars that has attained, or justly deserved, a higher reputa- 

 tion than the Pulkowa Catalogue. The observations on which 

 this catalogue is founded were made by Schweizer, Fuss, Linds 

 hagen, and Wagner, at the Pulkowa observatory between 1842 

 and 1853. The observations were reduced by the several ob- 

 servers, thoroughly discussed by Wagner, and published in 1869. 

 Only one observer was employed at any period. As these result- 

 have received high praise for their accuracy, and for their free- 

 dom from systematic errors, it may be of .■■ome interest to consider 



