NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: How Many Chronometers?
From: Frank Reed
Date: 2009 May 12, 13:37 -0700
From: Frank Reed
Date: 2009 May 12, 13:37 -0700
Henry Halbot, you wrote: "Although again probably past the prime era of the method, it reinforces my often stated opinion that, given more time, innovation, and practice, it would have been significantly perfected in its accuracy and consistency. " Well, the methods described in that article were not new, and I don't think they represent any approach to perfection. In fact, I see rather the opposite: advancing decadence. I read this article some three or four years ago, and if I remember correctly, someone wrote a follow-up letter to the MNRAS pointing out that this was a known method for rating chronometers --known for decades. Indeed, a decade earlier, one T.N. Were related his experiences rating his chronometer by lunars and even then he referred to lunars in that decade as "too much neglected". The interesting change in the time elapsed between those two articles is that Were's results were published in the "Nautical Magazine" while Toynbee's results were relegated to the MNRAS --"Monthly Notices of the Royal Astronomical Society". I say "relegated" because the Nautical Magazine was a publication aimed at practicing navigators while the Monthly Notices were targeted to astronomers. The method had gone obsolescent, and its remaining advocates were primarily rediscovering aspects that were relatively common knowledge in the heyday of lunars. By the way, I remember three or four follow-up letters to the MNRAS regarding Toynbee's article, so at least the astronomers enjoyed it! And you wrote: "We are also not advised as to whether or not the Captain had access to the newly calculated Lunar Tables which came out about the time of the published observations - I know only that the American Almanac first contained the new tables in 1855, but that does not necessarily mean that they might not have been available elsewhere before that date." I consider it highly unlikely that he had any updated tables. But be advised that there are two separate improvements here. The improvement which was made in the American Ephemeris (and which somewhat surprised the Europeans) was due to a very simple revision. They took the established model of the Moon's motion, which had been in use since the early 19th century, reduced more recent observations, and re-generated the basic tables (for a modern analogy, they effectively re-calculated the "coefficients of the time series" used in the algorithms for calculating the hourly positions of the Moon). This had not been done in Britain in decades so the errors in the predicted positions were slowly building up. If lunars had mattered to any significant degree to British shipping at this point, they easily could have done the same thing. If you examine the first American Ephemeris and compare the lunar distance tables with the standard British Nautical Almanac for the same year, the American tables are moderately more accurate. The average errors are something like 30% smaller. When I compared these a few years ago, I found that one could get even better results, surprisingly, by averaging the British and American predictions. Just a few years after the first release of the American Ephemeris, the models of the Moon's motion were significantly improved, and these improvements were rolled out in later years in the various almanacs in Europe and the US. But it was too little, too late. By 1860, there were very few practitioners of lunars at sea. I've posted it before, but for reference, here is T.N. Were writing about his experience with lunars in 1842: "CHRONOMETERS AND LUNARS. [After a preamble offering some revised positions for islands and points in the Indies...] I would now beg to offer some remarks relative to chronometers, and the fearful consequences that may arise if we place too much confidence in them, which from their beautiful and improved construction we are now too apt to do, and neglect those observations of the Heavenly bodies (which can only be of use by their being in constant practice) when you may have confidence in them. On my outward voyage, I had a beautiful watch of Murray's, which differed only nine seconds, in my run from Portsmouth to the Great Ladrone (on the coast of China). I placed great reliance on this watch during my homeward voyage, after leaving the coast of Java, and as soon as the moon came in distance I obtained a few sets of sights which gave the watch a considerable error, nearly three minutes. At this time I imagined my distances must have been incorrect, but the day previous to getting on the L'Agulhas bank, and in a run of about thirty days from Java Head, I was fortunate enough to get the mean of some thirty or forty distances, and I was much astonished that they gave the watch an error of ninety miles to the westward. I also carefully observed the sun's semi-diameter, and this corresponding with the Nautical Almanac, gave me confidence in my sights, and shook my confidence in the chronometers. It also fortunately placed me on my guard, and as on rounding the land, I was in a position to make Cape L'Agulhas at daylight I did not bear away. I was running nine knots at the time, with a strong S.S.W. wind, and did not see the Cape until 2 P.M., having run since daylight about 80 miles departure, which made the lunar sights as near correct as I could take the bearings of the land, at about four leagues distance in hazy weather. Had I been by chronometer, near the longitude of the Cape in the evening, I certainly should have borne away before morning, and had I done so, the melancholy fate of the ship which I commanded would, I fear, have been similar to the Arniston, Northumberland, and various others. How to account for this error in the watch has quite bewildered me, excepting it arose from magnetic attraction, as from some cause or carelessness a brace of pistols was placed close to the watches, which I had removed, directly I found the watches had altered their rates. The chronometer has since remained stationary, and retains the same rate as when I was off the Cape. On my passage towards England, after rounding the Cape, I took every opportunity of obtaining lunar distances ; and two days before I reached St. Helena, I spoke a ship which had also seen Cape L'Agulhas, and we differed seventy miles in our longitude. This startled me. But when I found my brother mariner had not taken any lunar sights, and having confidence in my instrument, I steered boldly for the island, and made it ahead about an hour before daylight. I was at St. Helena two days. Each day I got the Greenwich time from the observatory, and in taking the mean of my lunars about 100 distances, I was much pleased to find we only differed one second from each other. I would from these circumstances caution mariners, not to be too confident in their chronometers, and to lose no opportunity of obtaining lunar observations when practicable, and which I fear is now too much neglected, and which unless they are constantly practised can be of little avail, and can give no confidence to the observer. I have generally rated my chronometers by lunars, and have hitherto been fortunately correct in doing so, as they seldom retain the rate given by their constructors, and having two sextants, one of which being Troughton's, which as an instrument of confidence for navigation, is worth to me all the chronometers that were ever constructed. I have once before made some small contribution to your work, and am a subscriber from its commencement, and many an agreeable hour, I spend at sea in conning over its pages. Should yon think this worth inserting, or any part of it in your publication, I beg you will do so. I have the honour, &c., T. N. WERE, Commander of the Ship City of Derry. " -FER --~--~---------~--~----~------------~-------~--~----~ Navigation List archive: www.fer3.com/arc To post, email NavList@fer3.com To , email NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---