owing to these causes more than to any other, that the ancient method of computation from the Western Islands has been discontinued.

The other reason, which is very gravely urged for the proposed change of the first meridian, is, that to count the longitude from an English observatory is "a sort of degrading and unnecessary dependence on a foreign nation" and an "incumbrance unworthy the freedom and sovereignty of the American people."

Language of this sort really appears to us to be wholly unworthy of a serious answer: It is an address rather to the passions than the reason of our countrymen; and is an example of the introduction of national prejudices in matters of science, which cannot be too much reprobated.

It is no small objection to the proposed method, that the seat of government is far from being stable. Within a short time it has been at New-York, Philadelphia and Washington, and in the last congress a powerful attempt was made to remove it back to Philadelphia. The great increase of territory and population, in the southern and western parts of the union, may, in the course of a few years, occasion its removal to a different situation; and then, in conformity with the proposed plan, another new meridian must be adopted. The tendency of such changes is to increase the "confusion and difficulty," render the calculations much more complex, augment the labour, and cause mistakes and embarrassments in quoting the longitudes of places; and all this without being attended with one real advantage.

For these reasons it appears to us, that it would be altogether inexpedient, as well as extremely difficult, to bring the proposed method of computation into general use.

The most effectual way of doing it, would be to erect a national observatory at Washington, and calculate and publish annually a National Almanack adapted to that meridian; and the erection of such an observatory, under his direction, was perhaps one of the objects Mr. Lambert had in view. For the instruments actually necessary to determine the longitude of a place, with great accuracy, are not so very expensive as to require the aid of the nation to procure them; nothing more being requisite, to observe eclipses of the sun or occultations of the fixed stars by the moon, which are the best methods of obtaining the longitude, than a good telescope, a

clock or chronometer, and a quadrant or repeating circle; all of which might be purchased for four or five hundred dollars; and the same instruments might be used in observing eclipses of the moon or satellites of Jupiter, transits of Mercury or Venus, or the angular distances of the moon from the sun or a star. If the object of Mr. Lambert is to procure an observatory and furnish it with instruments necessary to make it respectable, the expense would be very great; and though we should not condemn that plan, we should much prefer to have the same sum appropriated in a more useful manner. And in our opinion it would be abundantly more beneficial to the country to expend it in obtaining a complete and accurate survey of the sea coast of the United States, for the want of which, our navigators are daily suffering in approaching our

shores.

This subject was introduced into the national legislature a few years ago, and we have understood that a considerable sum was appropriated for the purpose; but we have not yet heard of any part of the work being done, except the shore of North-Carolina, by Mess. Cole and Price; and we are inclined to believe that it is now discontinued; which is very much to be regretted, as the necessity of making such a survey is obvious to all who have ever used the present charts of the

coast.

We are so strongly impressed with the importance of this undertaking, that we cannot refrain from urging it, as we have on former occasions, and mentioning several facts that have come to our knowledge, though they are in some measure foreign from the subject of the present review. In the above survey of North-Carolina, the whole coast from Cape Hatteras to Cape Fear was found to be a third of a degree too far to the westward on the best charts and books extant; and what is far more important, the latitudes of some of the extreme shoals of that coast were laid down too far to the south by 10 or 12 minutes. From some observations made at Charleston, S. C. published in the National Intelligencer of Nov. 1, 1809, it appears, that in the most approved books and maps, the longitude of that place is given with nearly the same errour as that in the coast of North-Carolina; and by the late Spanish surveys, a similar mistake has been found in the situation of the eastern shore of the coast of Florida. These circumstances make it appear highly probable that like

errours exist in the whole shore of South-Carolina and Georgia, and prove fully the necessity of a more accurate survey of that part of the United States.

Mistakes of a different kind exist in the charts of the northern section of the union, from Martha's Vineyard to Cape Sable; where the extreme points of the coast and the adjacent islands are in general laid down with sufficient accuracy for most nautical purposes, in the survey made by Holland and his associates; but the numerous banks and shoals, extending many leagues from the coast, have never been explored with any tolerable degree of accuracy. Thus the shoal of St. George's bank, which is a very dangerous place in bad wea ther, is probably not marked accurately in any book or chart extant. The distance of this shoal from the light-houses on Thatcher's Island is 35 leagues by Holland's chart, whereas the real distance is nearly 45 leagues, as is well known to many seafaring people in Massachusetts. The longitude of the centre of the shoal is by the same chart 68° 34 min. which is nearly a degree too great. In the chart of Nantucket &c. published by Paul Pinkham, the distance of this shoal from the main land of Cape Cod is supposed to be rather more correct, but the longitude 661° is too small by about a degree; notwithstanding this errour the chart is probably as good for sailing from Martha's Vineyard to St. George's bank as any extant; indeed it appears to have been the standard to which compilers of charts have lately referred for this part of the coast; most of those shoals having been neglected by Holland.

These, and many other facts that might be mentioned, shew the necessity of taking a new and accurate survey. In doing this it would be necessary to take a complete body of soundings throughout the whole coast, to assist the mariner in determining the situation of his vessel, in foggy weather, or in the night, by the depth of water and the quality of the ground. A survey made in this manner from Martha's Vineyard to Cape Sable, including the whole bay of Massachusetts, would be of incalculable service in approaching that part of the coast, particularly in the winter, when the navigation is by far the most dangerous.

This work would require much time, labour, and expense, but when once well done, it would probably not require any material correction for many years. On the other hand, if an imperfect survey be made, it will answer no valuable pur

pose, as no confidence will be placed in it, and the whole labour will be lost. The survey of Massachusetts made twelve or fifteen years ago, by order of the state legislature, is an instance of this kind. This was entrusted to many persons, and some parts were so very imperfectly done, that the map abounds with errours, and as a chart of the coast it is wholly useless. The same labour and expense, properly applied, would have made an accurate survey of a smaller district, and the real service rendered to the community by this measure would have been much greater.

The failure in this survey was owing in a great degree to not having a person to superintend and direct the whole work. This point ought to be particularly attended to in a general survey. Perhaps it might not be amiss to divide the whole into three or four districts. The New England states might form one, extending from Cape Sable to Martha's Vineyard or Newport; the middle states, another; the southern states, a third; and the Louisiana territory, a fourth each of these districts being sufficiently extensive for any one person to su perintend. To render the chart complete it would be necessary to take in the coast of Florida and the northern parts of the Bahama Bank, which might be copied from the Spanish surveys with some corrections and additions.

The latitudes of the points of the coast might easily be found by a circle of reflection, but it would be much more difficult to ascertain the longitudes to a sufficient degree of precision. The committee in their report suggest the propriety of making use of the eclipses of Jupiter's satellites and the angular distances of the moon from the sun or a fixed star; but we are inclined to believe that it would conduce to the accuracy of the survey to make use chiefly of eclipses of the sun and occultations of the fixed stars by the moon, where the difference of meridians cannot conveniently be found by a trigonometrical survey. An eclipse of the sun observed under favourable circumstances will give the longitude within a few seconds of time, and this is the best method of finding it. An occultation of a fixed star carefully observed comes next in point of accuracy. These observations ought to be compared with others made at Greenwich, or at a place whose longitude from Greenwich is accurately known; by this means the errours arising from the inaccuracy of the moon's place in the nautical almanack will be avoided this

cause would have produced an errour of nearly half a degree in the observations made in the total eclipse of the sun of June 16, 1806. This method of correcting the errours of the tables is not followed by Mr. Lambert in calculating the obser vations of n Pleiadum, so that in the longitude he has assigned to the city of Washington, there may be a considerable errour arising from this source, particularly as Burg's tables (which are by far the most accurate of any extant) were not used.

The uncertainty of the longitude deduced from observations of the eclipses of Jupiter's satellites may be seen, by comparing above two hundred observations made by Mr. Goldingham, at Madras, between the years 1794 and 1802, and published in the Transactions of the Royal Society of London for 1808, Part 2. For the longitude of Madras, deduced from these observations, by comparing the observed times with those in the Nautical Almanack, varied in the extreme observations of the first satelite above 2°, in those of the second above 14o, and in those of the third about 3°. These differences in some of the observations he attributed to the haziness of the weather, the proximity of Jupiter to the horizon or to the moon ; to the twilight, or great light of the moon; or to the errours of the tables. This last source of errour might be obviated by making corresponding observations under a known meri dian, but Mr. Goldingham observes that "it must not be supposed that even this will give a correct difference of longitude, unless observed at both places under the same favourable circumstances, and with telescopes of the same powers." The method of determining the longitude by the angular distances of the moon from the sun or a fixed star, as usually practised at sea, would not be sufficiently accurate to find the longitude where very great precision is required, on account of the errours of the lunar tables.

This difficulty might, however, be obviated, and the method employed, by taking simultaneous observations of the angular distances, at a place whose longitude is accurately known. Thus for finding the longitudes of the head-lands or shoals of the coast of New England, observations might be made at Cambridge, the longitude of which was determined

It is possible that these extreme eours might be somewhat diminished by using the new tables of De Lambre.