Page:The American Cyclopædia (1879) Volume X.djvu/634

 628 LONGITUDE LONGMAN they show the distance of the moon from the sun and stars within 15" of a degree, answer- ing to about 7' of longitude, after allowing half a degree for errors of observation and under certain restrictions, and after comparison with astronomical observations for a period of 18 years, during which the lunar irregularities are supposed to be completed. The same rewards were likewise offered to any person who should discover any other method of determining the longitude at sea with the accuracy above men- tioned. At that period timekeepers were ex- pensive and less to be depended on than at the present day; they could easily be consulted, bat prudent mariners were cautious, and the motion of the heavenly bodies was more to be trusted. The eclipses of Jupiter's satellites, al- though answering well on land for determining longitude, could not be observed at sea on ac- count of the ship's motion, and the more prac- tical method of observing lunar distances for this purpose was adopted, Dr. Maskelyne, the astronomer royal, being the first to introduce them into use. The difficulties which had at- tended this method when first suggested were now removed. Prof. Mayer of Gottingen had formed lunar tables sufficiently correct to in- duce the commissioners of longitude to promote their practical application by the annual publi- cation of the nautical almanac, commencing in 1 767, and proper instruments for observing lu- nar distances had come into general use. The method of ascertaining the longitude at sea by lunar observations was followed until the great perfection with which chronometers were made caused it to be neglected. The last discovered and most accurate of all methods of determin- ing differences of longitude is by the use of the electro-magnetic telegraph. Captain (after- ward Admiral) Charles Wilkes, of the United States navy, is entitled to the credit of having made the first attempt in that way, soon after the wires were placed between Washington and Baltimore ; since that time the method of recording transits on a chronographic register by means of a galvanic circuit has been intro- duced, and in connection with telegraphic wires enables remote observers to record transits of the heavenly bodies simultaneously and with a degree of accuracy almost incredible. This method is called in Europe the American, but it may with greater propriety be called the United States coast survey method, as it had its origin and was perfected in that service. The application of this method has been also extremely important as a means of testing the correctness of the various astronomical meth- ods which have been used, or may still have to be used, for regions with which telegraphic communication is impossible or inconvenient. Prof. Asaph Hall of Washington makes the following remarks on this important subject : " The method of defining longitudes by moon culminations is so simple in theory and so easy of application, that when only an approximate value of the longitude is desired, this method will often be applied. In case, however, we wish an accurate determination of a geograph- ical position, such as may be necessary for a station occupied in observing the transit of Venus, it is well that we should not over- estimate the accuracy of this method. In many of the estimates that have been made we have an illustration of the existence of constant errors which render such estimates of accuracy wholly illusory. ... In nearly all the deter- minations of longitude by moon culminations, where a large number of culminations have been observed, the computed probable error of the result is only a small fraction of a second ; but the telegraphic determination of the same points shows errors in the old determinations of two, three, and even four seconds of time. Thus the longitude of San Francisco, deter- mined from 206 moon culminations, was found to be four seconds in error. The most decisive experiment on this point is, however, the de- termination of longitude between Europe and America. The three determinations of longi- tude between Greenwich and Washington by the United States coast survey, by means of the Atlantic cable, give the difference of lon- gitude 5 h. 8 m. 12*2 s. The following are the determinations of the same difference of lon- gitude by moon culminations : AUTHORITIES. No. culm. Longitude. Loomis 150 5h. 8m. 9'3s Gilliss. 894 5 8 lO'O Walker 5 8 9-6 Newcotnb 279 5 8 11-6 Newcomb. 163 5 8 9-8 showing errors 2*9 s., 2-2 s., 2-6., 0-6 s., and 2 -4 s. In his last determination, Prof. Newcomb used the observations of 1862 and 1863, when the transits both at Greenwich and Washington were recorded by the chrono- graphic method. It seems therefore fair to conclude that a longitude determined by moon culminations may be in error two or three seconds, even if we could use an infinite num- ber of observations." LONGLAND, Robert. See LANGLAND. LONGMAN. I. Thomas, an English publisher, born in Bristol in 1699, died in London, June 18, 1755. In 1716 he was apprenticed to John Osborn, a stationer and bookseller of London, with whom he entered into partnership in 1725. in Paternoster row, in the place still occupied in the same business by his successors. The chief publications in which he was concerned were Chambers's " Cyclopaedia of Arts and Sciences," the precursor of the different cyclo- pedias which have since appeared in Europe and America, and Johnson's "Dictionary," of which latter work he was one of the six ori- ginal proprietors. II. Thomas, nephew of the preceding, born in London in 1731, died at Hampstead, Feb. 5, 1797. He was taken into partnership by his uncle in 1754, and subse- quently entered into partnership with Mr. Rees,