Tuesday, September 16, 2014

What do the Numbers 3500 and 1542 Tell Us? Part One

On the introduction page, I described the discovery on Gardner Island of the skeletal remains of the castaway and, nearby, several items including a sextant box.  This is going to be the first of several posts about the sextant box, which is of central importance to the Nikumaroro Hypothesis.  Tighar hypothesizes that the sextant box once contained a sextant made in American for the U. S. Navy that was later acquired by Fred Noonan, Amelia Earhart’s navigator.  If the sextant box can be traced back to Noonan, it would be the long sought-after smoking gun that proves the Nikumaroro hypothesis.  Tighar has been unable to establish a definitive link between the sextant box and Noonan, but it has built an interesting circumstantial case linking the two.  In this post, I want to look at a single part of this circumstantial case — that the sextant box once held an American-made, U.S. Navy sextant. 

The whereabouts of the Nikumaroro sextant box, if it still exists, are unknown, so it can’t be examined for clues about its origins.  All we have to go on is the description of it given by Gerald Gallagher, the colonial official on Gardner Island when the sextant box was found, and remarks made by those who were asked to examine it after Gallagher sent it to Western Pacific High Commission (WPHC) headquarters in Fiji.  Fortunately, what these people had to say about the sextant box was documented in the files of the WPHC, which are now held at the National Archives of Kiribati; it was quite a coup for Tighar to bring these obscure documents to light.

The document in the WPHC files that provides the most important clue about the sextant box is the earliest one mentioning its existence — Gallagher’s message to WPHC officials informing them of the discovery of the castaway’s remains.  In part, this message reads:

  “Some months ago working party on Gardner discovered human skull – this was buried and I only recently heard about it. Thorough search has now produced more bones (including lower jaw) part of a shoe, a bottle, and a sextant box. … Sextant box has two numbers on it 3500 (stenciled) and 1542 – sextant being old fashioned and probably painted over with black enamel.”

D.C.M MacPherson, a colonial doctor who had been apprised of Gallagher’s message, wrote a memo that offered what seems in retrospect to have been a key insight:

Up till the present the number on the sextant case appears to afford the most hopeful means of identification. The instrument itself moreover, if a good one, should have engraved on it a number assigned either by the Bureau of Standards in the case of the United States, or the National Physical Laboratory in the case of the United Kingdom.  This number indicates as a rule the result of tests for which compensation requires to be made in using the instrument.

As we’ll see, MacPherson probably wasn’t far off the mark.  If WPHC officials had made inquiries with the proper authorities about the two sextant box numbers, they might have succeeded in tracing it to its last owner and perhaps would have identified the castaway.  But that didn’t happen; as explained in my introductory post, after the castaway’s bones and the sextant box were examined in Fiji, WPHC officials concluded that the castaway wasn’t Amelia Earhart (1), and no further effort was made to determine the identity of the castaway or the origin of the sextant box.

Tighar believes that the castaway was Earhart, and so when Gallagher’s message came to light it began searching for examples of sextants or sextant boxes that were marked with two serial numbers.  It was possible to find sextants made in the United States, England, France, and Germany marked with one four-digit serial numbers, but for several years no sextants having two serial numbers could be found, nor could any sextant boxes be found that were marked with even one serial number.  Eventually, Tighar discovered that U.S. Navy sextants made in the early twentieth century were marked with two serial numbers, one a U.S. Naval Observatory serial number, and the other a manufacturer’s serial number.  The Naval Observatory had the task of inspecting newly purchased U.S. Navy sextants for general suitability before they were put into service and re-inspecting them once in service; each sextant that passed its first inspection was assigned a Naval Observatory serial number, which was hand-etched into the sextant’s arc.  The manufacturer’s serial number was stamped into the sextant’s index arm or into its arc.

 Manufacturer's Serial Number, Brandis #3985
(photo eBay)
Naval Observatory Serial Number #4826 on Brandis Sextant # 3985
(photo eBay)
Tighar has identified fifty six U.S. Navy sextants and sextant boxes made before World War II that are marked with both Naval Observatory and manufacturer’s serial numbers and lists them on its web site (2).  Their Naval Observatory serial numbers range from 34 to 4954, forty-six of which have four digits.   Some of the sextant boxes are marked with the Naval Observatory serial number of the sextants they held, either stamped (i.e., mechanically impressed) into the box or appear on a medallion attached to the box.  All of the sextants on Tighar’s list were made in the United States, forty-five of them by Brandis & Sons, a Brooklyn based precision instrument manufacturer.  The other eleven sextants were made by three other instrument manufacturers — seven by Keuffel & Esser, three by Buff & Buff, and one by Warner & Swasey.

All forty five of the Brandis sextants have four-digit manufacturer’s serial numbers, which range from 1844 to 5953, and forty one of the Brandis sextants have four-digit Naval Observatory serial numbers to go with their four-digit Naval Observatory numbers.  Some of the Brandis sextant boxes are marked with both Naval Observatory and Brandis serial numbers, with the latter stamped on the box in ink.  As far as I can tell, none of the other manufacturers stamped their serial numbers onto their sextant boxes (3).

Sextant Box for Brandis #3985
(photo eBay)
So, U.S. Navy/Brandis sextant boxes fit Gallagher’s description of the Nikumaroro sextant box pretty well: they’re marked with two four-digit numbers, one of which is stenciled onto the sextant box.  If that is what the Nikumaroro sextant box was, then the stenciled number — 3500 — was its Brandis serial number, and 1542 was its Naval Observatory serial number.  How does this pair of serial numbers compare to those of actual U.S. Navy Brandis sextants?  Table 1 indicates the Naval Observatory and Brandis serial numbers of twelve sextants from Tighar’s list whose Naval Observatory numbers are in the vicinity of 1542, the putative Naval Observatory serial number of the Nikumaroro sextant box.

Broadly speaking, the Brandis serial numbers in Table 1 tend to increase as the Naval Observatory serial numbers do, but if there is a well defined mathematical relationship of some sort between the two sets of numbers, I don’t see it.  The Naval Observatory must have been issuing serial numbers in strict numerical order, but the Brandis serial numbers don’t monotonically increase with increasing Naval Observatory number, so apparently the Brandis sextants weren’t put in order of their Brandis serial numbers before they were inspected and assigned their Naval Observatory serial numbers.  While there doesn’t seem to be an obvious relationship between the two sets of serial numbers, we can nevertheless see that around the time that the Naval Observatory assigned serial number 1542 to an instrument, it was inspecting Brandis sextants with manufacturer’s serial numbers in the 3300 to 3900 range.  So, the 1542/3500 serial number pair of the Nikumaroro sextant box is indeed a plausible one for a US Navy Brandis sextant.

Could the Nikumaroro sextant box have held a US Navy sextant made by Keuffel & Esser, Buff & Buff, or Warner & Swasey?  Table 2 indicates the manufacturer’s and Naval Observatory serial number pairs of the non-Brandis sextants on Tighar’s list; none of these number pairs consist of two four-digit numbers, and I certainly don’t see a correlation between the 1542/3500 number pair and any of these number pairs, so I tend to think the Nikumaroro sextant box could not have been held one of these manufacturers’ sextants.

Sextant marked with asterisk is not on Tighar's list; see (4)

Why try to make inferences based upon the serial numbers of surviving examples of sextants and sextant boxes?  Can’t we simply check the Naval Observatory’s sextant inspection records to determine whether a sextant having the 3500/1542 serial number pair actually existed?  Tighar has tried to do so, but unfortunately the Naval Observatory no longer has records of its sextant inspection work; Tighar has searched for these records in the U.S. National Archives but so far has been unable to find them.

While we lack definitive proof that the Nikumaroro sextant box once held a U.S. Navy Brandis sextant, the circumstantial case for this hypothesis is nevertheless a strong one.  U.S. Navy Brandis sextant boxes fit Gallagher’s description of the Nikumaroro sextant box well.  They are, at least in some cases, marked with two four-digit numbers, one of which might be said to be stenciled, and the two numbers on Nikumaroro sextant box, 3500 and 1542, appear to be a plausible number pair for such a sextant box.  No other sextant boxes we know of come nearly as close to matching Gallagher’s description.  I’m not sure I’d want to bet the farm on this hypothesis being right, but I definitely wouldn’t bet the farm that it is wrong.  As strong as this case is, there is one aspect of Gallagher’s description of the Nikumaroro sextant box that can’t be easily reconciled with it being a U.S. Navy Brandis sextant box, and that suggests that maybe—just maybe—the Nikumaroro sextant box was something else.  This I will discuss in a future post, but the hypothesis that Nikumaroro sextant box once held a Brandis sextant made for the US Navy is a sound one, so in my next post, I’ll assume that indeed is what the Nikumaroro sextant box was, and I will look at what the numbers 3500 and 1542 tell us about when a sextant with those numbers would have made by Brandis and inspected at the Naval Observatory.

Comments, corrections, additional relevant facts, differing viewpoints, etc., are always welcome (no one will be banned, blocked, or castigated for offering differing opinions).  Send to gardnersghost@gmail.com 
1)  The doctor who examined the skeletal remains concluded that the remains were those of a short, stocky male, thus excluding the possibility that the castaway was Amelia Earhart.  It is not clear why WPHC officials didn’t consider that the castaway might have been Fred Noonan; perhaps they thought Earhart was flying solo.
2)  http://tighar.org/wiki/Sextant_box_found_on_Nikumaroro
3)  However, the sole Warner and Swasey sextant box on Tighar's list has the number '69' as part of the larger ‘W&S HG Sex 69’ marking stenciled onto the box’s lid; ‘HG’ indicates it was a high grade, rather than surveying sextant, and 69 was the W&S serial number. If Gallagher had seen more than a number I think it is safe to say he would have said so.
4)  Found on liveauctionneers web site; see photo on the supporting materials page.