Although the French mathematician Lagrange discovered the equilibrium solutions for a gravitational system of three bodies as long ago as the 1770s, it was not until 1906 that the first astronomical observations were made of an asteroid that illustrates the stability of such behavior when it, Jupiter and the sun are located near the vertices of an equilaterial triangle. Right?
Wrong! The Heidelberg astronomer Max Wolf indeed discovered the asteroid TG, later known as (588) Achilles, in February 1906, and within a matter of weeks, the significance of the fact that it preceded Jupiter in its orbit by some 55 degrees was illustrated by Carl Charlier of Lund. In October of the same year, (617) Patroclus was discovered, this one then 57 degrees following Jupiter in its orbit. Collectively known as Trojans and named for heroes on both the Greek and Trojan sides in the ancient war, further such objects discovered throughout the century have fully confirmed Charlier's interpretion. With the publication of the latest batch of Minor Planet Circulars earlier this week, as many as 170 Trojans had been given permanent numbers (a point that taxes the efforts of those who search the pages of The Iliad and other works for suitable names!). There are also 113 Trojans that have been observed at two or more oppositions and will surely qualify for numbering and naming in the future.
In what way is the above claim wrong? It is wrong because an observation of a Trojan has now been recognized from 1904! This happened on Sept. 26 when Gareth Williams, Associate Director of the Minor Planet Center, was examining the orbit of 1999 RM11, a recent discovery from the LINEAR (Lincoln Laboratory Near Earth Asteroid Research) project. In a collaboration with this project, he had first linked together observations of some 4600 previously unknown asteroids in multiple-night data obtained with the LINEAR telescope in New Mexico during the second week in September. This particular object, recorded by LINEAR on four different nights, was one of a handful of new Trojans that Williams recognized in the data. He was then able to identify observations from single nights of data from the Spacewatch and LONEOS near-earth-object projects on Sept. 3. After bootstrapping this 10-day orbit calculation back to find two more nights of LONEOS observations in early August, the improved orbit calculation was good enough to reveal that the same object had been recorded by NEAT, Eleanor Helin's near-earth-object project at the Jet Propulsion Laboratory, on two consecutive nights (again linked in collaboration with Williams) in April 1996, when it had been designated 1996 HJ22. Williams also realized that it had been photographed during Helin's program with the 0.46-m Schmidt at Palomar in November 1978, when it was called 1978 VH6. On linking the 1999, 1996 and 1978 observations together, it was then trivial for him to complete the job and identify single-night observations of the object in 1988, 1987, 1985--and 1904.
So what was this 1904 observation? Interestingly, special attention was drawn to it at the time, because the observer was looking for a specific slow-moving object. The observation was in fact made by the celebrated observer Edward Emerson Barnard with what was then the largest working telescope in the world--the 1-m refractor at the Yerkes Observatory. Barnard was looking for Phoebe, the ninth satellite of Saturn, discovered six years earlier, but for which the orbit determination was still unclear. Discoverer William H. Pickering had provided Barnard with an ephemeris for the satellite, and on the evening of Sept. 12, Barnard detected an object close to this prediction. He estimated the magnitude as 16.7 and established that it moved to the west over the course of some 25 minutes and possibly slightly to the south.
Later, when Pickering had been able to do a more complete analysis of the Phoebe data, he pointed out that Barnard's object was 2 arcmin from Phoebe's position and suggested it was a star. In a paper he wrote about his observation for the Astronomische Nachrichten in 1907, Barnard quite indignantly pointed out that "no one at all familiar with micrometer work with a large telescope would for one moment question the fact that this object was moving", and furthermore, he verified that there was no star at the position on a photograph of the region obtained by Pickering. Suggesting that the motion was probably too great for it to be a satellite of Saturn anyway, Barnard remarked that the object had to be an asteroid, but that "If an asteroid, it is doubtless lost for good". Nevertheless, he added that "Some time the position of this faint asteroid may be important and I wish to put the observations on record". The object was therefore interpolated into the asteroid designation system of the time as OVa, which in the modern system was changed to A904 RD.
Although Barnard dismissed the possibility that his object was an unknown Saturnian satellite, was he aware that it was in fact moving more slowly than other asteroids known at the time and thus likely to be farther away? He measured the westerly motion in right ascension to be 0.226 arcsec per minute. The computed value for 1999 RM11 at the time would have been 0.222 arcsec. (The declination motion was 0.042 arcsec per minute to the south, but Barnard's declination observations relative to the comparison star covered only 5 minutes and he wisely chose not to supply a figure.) Of course, since the object was near the meridian when Barnard observed it around 10 p.m. local time, the slow motion of a Trojan would not be too obvious, although it would have been only some two-thirds of that of a main-belt asteroid. Given his consideration at the time that he was observing Phoebe, it is difficult to fault Barnard for not recognizing that he had instead made the earliest observation of an object belonging to a new dynamical class; if he had, he would presumably have made an effort to follow up the object on subsequent nights. As it was, he thought he had also observed Phoebe near Pickering's ephemeris a month earlier, although he later dismissed that observation as spurious. However, even in his 1907 paper, written when the discovery of three Trojans had been established, he made no hint at the possibility he had glimpsed--and lost--a fourth.
Could Barnard's 1904 observation have been explained prior to this careful reconstruction by Williams just 95 years later? That would have been very difficult, though perhaps not entirely impossible to do from the published data. In 1978 the Palomar observations were made on three of four consecutive nights, and an orbit computation shows that the object is indeed a Trojan. That computation was made--again by Williams--in 1996 and published a couple of weeks before the 1996 observations were made. The fact that the 1978-1996 linkage was not made at the time illustrates the current general (though not always complete) intractability of the identification problem when a moderately good orbit is lacking at any of the oppositions.
The 1904 observation is perhaps not "important" in the sense that Barnard meant, because it really does not add anything to the determination of 1999 RM11's orbit. A calculation covering the 1975-1999 data "predicts" it within 3 arcsec, which is as well as it fits if it is included in the computation with unit weight. In fact, the single-night observations in 1975, 1985, 1987 and 1988 are also essentially unnecessary for the orbit determination, in that an orbit from the 1978, 1996 and 1999 data alone represents them within their likely uncertainties; in this case the 1904 residual increases to 4 arcsec, still acceptable for single-star micrometry, and with a calculated visual magnitude of 16.5 from the modern data, showing that there was no doubt about the object's identity.
The only importance to Barnard's observation of 1904 is historical: a suggestion of what might have been. Barnard could have been the discoverer of the first Trojan in the group following Jupiter. A year and a half would still then elapse before the discovery of the first member of the preceding group.
Brian G. Marsden
1999 October 1
This page may not be copied onto other Web sites, but other sites may place "clickable" pointers to this page. Photo-copies of this information sheet may be distributed for educational purposes, provided that no charge is made in doing so; such photocopies must include full credits. Journalists and writers may quote from this "information sheet", provided that proper credit is given. Further information is available by writing to Mail Stop 18; Smithsonian Observatory; 60 Garden St.; Cambridge, MA 02138; U.S.A.
Back to index of Press Information Sheets.
Index to the CBAT/MPC/ICQ pages.