Whatever Became of "Snickers"?


"Snickers" was the unofficial name given to a possible dwarf-galaxy satellite of the Milky Way that is seen only in the 21-cm radio emission line of hydrogen. The discovery paper appeared in the November 1, 1975 issue of the "Astrophysical Journal Letters" entitled "A New Milky Way Satellite Found in 21-centimeter Line Observations" by S. Christian Simonson III, then at the University of Maryland (Simonson, S.C. 1975, ApJ, 201, L103). This paper interprets a velocity anomaly in the hydrogen maps of the Galaxy in the direction of the constellation of Gemini as the tidally-distorted, rotating disk of a dwarf galaxy at a distance of about 17 kiloparsecs from the sun. This distance puts it substantially closer to the Milky Way than the Large and Small Magellanic Clouds, which lie at distances of about 52 and 63 kpc, respectively. No counterpart to this proposed new galaxy has been detected at any other wavelength, but its proximity to the Galactic plane puts it along a line of sight that is substantially obscurred by dust.

The first appearance of the name "Snickers" is in a "Science News" article announcing the discovery entitled "Looking for Mr. Goodgalaxy" (1975, Science News, 108, 309). The name is attributed to "[a]ssociates of Simonson [who] have named the new galaxy Snickers because it is peanuts compared to the Milky Way." There are only two other places I can find this name in print: a brief mention in the article on the Large Magellanic Cloud in the Volume 2 of Burnham's "Celestial Handbook" (pg 837 of the 1977 Dover edition), and in the table of "Galaxies with Proper Names" that was printed in the 1989 and 1990 issues of the "Observer's Handbook" of the Royal Astronomical Society of Canada (but it was absent from this table before and after these dates). It is not so named in a News Notes article which appeared in the April 1976 issue of "Sky & Telescope," nor does this name appear in any of the scientific literature on the subject. Given that the name is meant to be light hearted, this is not surprising.

My difficulty finding references to the popular name of this galaxy, and even greater difficulty finding any mention of it at all in scholarly or popular reviews written about the Local Group since the discovery paper in 1975, led me to wonder what really happened to "Snickers." Starting with Simonson's 1975 ApJ Letter, I began my own search for "Snickers" by consulting the "Science Citation Index" and the "Astronomy & Astrophysics Abstracts" for all years since 1975, and then following possible leads through review articles, conference proceedings, including tracing the literature surrounding "Weaver's Jet," the other name given to this velocity anomaly.

For those who don't wish to work their way through this merry little paper chase, the bottom line of the story is that the original interpretation of the hydrogen velocity anomaly in Gemini as due to a nearby, tidally distorted satellite galaxy seems to be incorrect. There is no single, damning disproof, just the weight of many little facts and inconsistencies which in my opinion overwhelms Simonson's interpretation. Since 1975 we have come to learn that the interstellar medium is far more complex and dynamic than had been previously suspected, and the velocity anomaly seen here is but one of many such features associated with the gaseous disk of our galaxy. The case for "Snickers" was never terribly airtight, and the subsequent literature points to the more significant conceptual weaknesses. One fact that struck me early in my research is the near total absence from the very beginning of any discussion of Snickers in the Local Group literature. Snickers is absent from tables of the members of the Local Group (e.g., Table 1 in Hodge's 1989 review of stellar populations in the Local Group; Hodge, P. 1989, ARA&A, 27, 139), and I have not even encountered a Local Group paper that discusses Snickers to discount it. One interpretation of this silence is that Simonson's interpretation has not been accepted by the Local Group community, though there is nothing in print in either refereed journals or conference proceedings to say why explicitly. This absense of citation, for or against, made the search that much more difficult.

For those looking for a juicy tale of scientific controversy, I'm sorry to disappoint. While it made for cute copy in the newspapers, it seems to have generated a rather sparse paper trail through the literature. There are no ringing praises, stinging damnations, or bitter exchanges of ApJ Letters. In fact, there is precious little response at all! The idea seems to just fade away, slowly sinking under the combined weight of new observations and changing perspectives about the nature of the interstellar medium.

"Looking for Mr. Goodgalaxy"

Our tale begins with an article in 1975 Nov 1 issue of the Astrophysical Journal Letters by Dr. S. Christian Simonson III, then at the University of Maryland. The paper (Simonson 1975, ApJ, 201, L103) is entitled: "A New Milky Way Satellite Found in 21-centimeter Line Observations." It describes an anomalous velocity feature in 21-cm maps of the Galaxy published by Weaver & Williams in 1973 and 1974 (1973 A&AS, 8, 1; 1974 A&AS, 17, 1 & 251; also Weaver 1974 in "Galactic Radio Astronomy", IAU Symposium 106, eds. F.J. Kerr & S.C. Simonson III, 573) in the direction of the Orion-Gemini region of the galaxy, a region generally referred to as the "galactic anticenter."

To digress briefly and give some background for the non-specialists in the audience, we trace neutral hydrogen gas in the interstellar medium by its emission of a bright emission line at a wavelength of 21-centimeters in the radio region of the spectrum. This line arises from hyperfine electron spin-flip transition in the ground state of neutral atomic hydrogen. In standard technical usage, we designate neutral atomic hydrogen as "HI" (read: "H one") for short. For the most part, the HI gas in our galaxy, as traced by the 21-cm line, is confined to a thin disk with concentrations of massive HI clouds that appear to outline spiral arms. All of these clouds have a large-scale pattern of line-of-sight velocities relative to earth (measured by the Doppler shift of the line) consistent with the expected pattern of Galactic rotation. This pattern is complicated due to the fact that we are riding around with the rest of the Galaxy, so our reference frame is less than ideal. Most of the observed gas shows velocities consistent with the simple motions of gas clouds orbiting in a flattened disk. A few regions, however, have what are called "anomalous" velocities. In 1973, in a survey of 21cm emission in the direction away from the galactic center, Weaver & Williams observed "high velocity" clouds or "streams of clouds" whose motions significantly deviated from the local pattern of Galactic rotation. It is one of these high-velocity clouds with an "anomalous" blueshift that is the subject of Dr. Simonson's paper.

Weaver & Williams identified a region of unusual negative (blueshifted) velocities, about -87 km/sec with respect to our local standard of rest, in the direction of the Orion-Gemini region. In later work, one sometimes finds this particular feature referred to as "Weaver's Jet". Given that the local rotation speed is about 200 km/sec in our part of the Galaxy, this kind of "peculiar" motion stands out. Whatever the nature of this region, its velocities do not share those of the surrounding disk material (seen overlapping in projection), and so it defines a kinematical "subsystem" in the HI distribution. The interpretation of HI features that do not strictly follow Galactic rotation is complicated, as we cannot directly measure the distances towards HI clouds; only the distribution of HI gas projected onto the sky and the velocities of those clouds along our line-of-sight are actually observed. If one assumes a particular Galactic rotation model, then it is possible to derive distances to clouds with a given velocity and position on the sky, but only if the motions are assumed to obey the simple rotation model (see "Galactic Astronomy" by Mihalas & Binney for the details of how this is done). For anomalous velocity features, like those in question here, one has to work harder to figure out what is being seen. Simonson describes this particular HI cloud in his paper as "...an HI feature near the galactic plane that appears to result from a small, nearby, tidally distorted galaxy."

Such a statement is loaded, since all we really have is a relatively low angular resolution HI 21-cm map. A better way of putting it might be to say that this feature is is "consistent with the gas distribution of a small, nearby...". This states the case more clearly. As Simonson further points out, there is no suggestion of an optical counterpart on the Palomar sky survey plates. The sole motivation for the identification of this feature with a companion galaxy is that it shows qualitative agreement with the results of a set of computer models that Simonson ran simulating the tidal distortion of a gas disk as it passed by our Galaxy. His two principal points were that (1) the velocity difference from the "normal" disk gas is larger than that seen in any other streaming gas features in the Galaxy, and (2) the lines have very flat profiles and there is a distinct velocity gradient across the feature, similar to what is seen along certain directions in M31. The first argument turns out to be weak, as at that time little was known about the systematics of the HI gas in our Galaxy. The second is just a restatement that there is rough qualitative agreement between the HI observations and his model calculations. From this, he concludes that the feature "arises from the systemic velocity and rotation of an external galaxy." With this interpretation in hand, he then estimates a mass and distance to the proposed galaxy using a mass model for our Galaxy. The reasoning is involved, but he comes up with a mass of about 1.0+/-0.2x10^8 solar masses of HI at a distance of 17+/-4 Kiloparsecs from the sun. This would make the putative galaxy a dwarf galaxy a little smaller than the Small Magellanic Cloud, and overall the nearest galaxy to our own among the various dwarf galaxies that orbit it.

On the whole, the paper is straightforwardly presented and very clear. The interpretation, however, is not too well backed up with observations, so the case for a nearby galaxy is very weak, but nonetheless suggestive.

There is no evidence for this galaxy other than the 21-cm line observations, and the interpretation given by Simonson is not the only one possible. He also mentions a suggestion by Weaver that the feature is either a "jet" or a part of a supernova remnant. Because of its proximity to the plane of our own galaxy, the line of sight to the proposed galaxy is heavily obscured by dust clouds. Because we are viewing this region through the plane of our own galaxy, the stars in the proposed dwarf galaxy would be mixed in and confused with a large number of foreground stars. Ideally distances would discriminate between foreground and dwarf galaxy stars, but since there is no a priori way to measure a distance to a given star, one has to be more clever. In this case, the proposal was made to search for distinctive RR Lyrae variable stars in this direction. These are pulsating giant stars that have a [mostly] known luminosity and distinctive pattern of variation. If you can find an RR Lyrae star and measure its apparent brightness and color (the latter to determine the amount of dust absorption between us and the star), in principle one can obtain a fairly accurate distance to the star. If that distance were large enough, it might confirm that the RR Lyrae star is in the proposed dwarf galaxy and not simply a foreground star in our Galaxy's disk. The Sky&Tel News Note mentioned that follow-up work was being planned, but a subsequent search of the literature has failed to turn up any signs that the plans were carried out.

While the response to the Simonson paper was essentially silence among astronomers who study the Local Group, it seemed to generate a fair amount of excitement among some gamma-ray astronomers, as the location of the feature on the sky lies near the estimated location of a reasonably bright gamma ray source called (in the parlance of the time) Gamma195+5. In a sequence of three papers by C.J. Cesarsky et al. (1976, A&A, 48, 481), G.F. Bignami et al. (1976, A&A, 51, 319, and 1977, A&A 54, 951), and D.J. Thompson et al. (1977, ApJ, 213, 379), the possible association between the two is entertained, then dismissed. The principal reasons cited for the dismissal of an association between the HI velocity anomaly and the gamma ray source are that there is (1) a significant mismatch in projected location between the two, and (2) the gamma ray intensities inferred from the distance to "Simonson's galaxy" (as it is called in these papers) are about a factor of 100 larger than expected from such an object. In addition, deep photographs made at a wavelength of 8000 Angstroms by Bignami et al. (1976) failed to find any obvious optical counterpart at the location of the proposed galaxy's nucleus. Since the line-of-sight towards the proposed galaxy does not pass through any dark clouds or regions of unusually high interstellar extinction (Av to this region is only 3-4 magnitudes), working at these long wavelengths should mitigate (but by no means alleviate) any of the usual problems with dust.

Further work touching on this unusual velocity feature by DeNoyer et al. (1977 ApJ, 213, 379) suggested that the "complex of low-latitude negative velocity clouds" was local, though their arguments are not all that conclusive. Cursory mention of Simonson's 1975 paper is found in a few other articles, but no comment on the interpretation of this feature as a companion galaxy is given. In general, it is mentioned only in passing, usually in introductory comments. What is most striking is that essentially all of the discussion of this proposed galaxy is found in the HI literature and the gamma-ray literature: no papers on the Local Group discuss the galaxy at all.

The most convincing evidence against Simonson's interpretation is found in a 1979 paper by W.B. Burton & R.L. Moore (1979, AJ, 84, 189). They note that in newer HI maps with substantially greater spatial resolution and sensitivity, the HI clouds with a large negative velocity are associated with disturbances in adjacent regions which themselves show "normal" velocities consistent with a rotating galactic disk. In particular, they observed secondary "high-velocity streams," and an intensity minimum that cannot be reconciled with the satellite galaxy model. They suggest instead that the feature identified by Simonson as a tidally distorted satellite galaxy is instead the focus of three HI gas streams.

R. Giovanelli (1980 ApJ, 238, 554) found another HI feature with anomalous velocities in the galactic anticenter region, showing that such features are not as uncommon as previously thought. T. Watanabe and collaborators (Watanabe 1982, A&A, 111, 333, and Watanabe et al. 1982, PASJ, 34, 249), suggested that their observations were consistent with the "Weaver Jet" originating at the collision of two giant expanding HI shells. The second paper reported the additional discovery of faint diffuse 2.6mm wavelength emission from CO molecular clouds around the position of the "Weaver Jet". However, as Heiles (1984, ApJS, 55, 585) points out, the HI structures observed by Watanabe may be unrelated because their systemic velocities relative to the local standard of rest are different that those in the "jet", though a connection cannot be ruled out on the basis of the current data.

This latter paper by Carl Heiles is entitled "HI Shells, Supershells, Shell-like Objects and Worms" in the interstellar HI distribution. "Worm" is Heiles' name for an expanding shell that has burst open on one end, appearing like the worm tubes seen in marine sediments. Heiles mixes no words; he completely dismisses Simonson's interpretation, making it the only open condemnation of the suggestion I can find in print. What is notable about the Heiles paper is its illustration that structures like those studied earlier by Weaver & Williams and Simonson are not all that uncommon in our Galaxy. All subsequent papers that mention the original Simonson paper do so with little or no comment, except perhaps to note that the result is "controversial." Again, comment on the work is confined primarily to the HI literature, and no mention appears at all among the literature dealing with the Local Group.

All of this adds up to suggest that Simonson's interpretation of the HI velocity anomaly in Orion-Gemini as due to a nearby, tidally distorted dwarf companion to our own galaxy is incorrect. We have since learned that such structures are not all that unusual in our Galaxy, and may be associated with energetic processes like supernovae or colliding interestellar clouds. As radio telescopes and interferometric techniques have improved in sensitivity over the years, the rich structure of the gaseous disk of our Galaxy has been revealed. It thus seems that "Snickers" is another of those ideas in astronomy that do not work out over time. I hasten to emphasize that the suggestion was not obviously wrong; it was fully consistent with the data and view of our Galaxy in 1975. The conclusion that I draw from this is that "Snickers" isn't. Further, I find the near total absence of discussion of this proposed satellite galaxy in the Local Group literature from the very beginning to be telling: it seems it was never accepted by that community, though there is nothing in print in either refereed journals or conference proceedings to say why explicitly.

While I am coming down on the side against Simonson's interpretation, the basic idea of searching for nearby or distant galaxies obscured by the plane of our own Galaxy using HI emission is not without merit. A recent demonstration may be found in the work of Frank Kerr & Patricia Henning at the University of Maryland (1987, ApJ, 320, L99). They have used improved 21-cm techniques to search for galaxies behind the Milky Way. The article cited above is a report of their pilot study, which found some 16 distant galaxies. The large scale patterns seen in the distant galaxy distribution (like the so-called "Great Wall") are seen to disappear behind the disk of our galaxy and pop out the other side, so the motivation for such a search is clear. Thus far, there are tantalizing suggestions, but no definitive results from these searches.


After writing the original article, Ken Croswell drew my attention to an abstract of a brief talk given by Dr. Simonson at the Berkeley AAS Meeting in Summer 1993 (Simonson, S.C. 1993, BAAS, 25, 892). It is clear from the abstract, and Dr. Croswell's report of the talk that Dr. Simonson, now at the Lawrence Livermore National Laboratory, still holds to his interpretation. It will be interesting to see if this public re-surfacing of Snickers from its relative obscurity leads to any new discussion.


This article was originally posted to the internet's USENet electronic bulletin board sci.astro in Spring of 1990, and has subsequently been substantially revised (and cleaned up). My thanks go to Steve Allen (Lick Observatory/UC Santa Cruz) and Pat Harrington (Dept. of Astronomy, U Maryland) for pointing me in the right direction at the start of my search, and I am most grateful to Ken Croswell who triggered the major revision by pointing out to me the Nov 1975 Science News article that gives the derivation of the name.

Journal Abbreviations used above:

  ApJ = Astrophysical Journal.  Page numbers beginning with "L" 
        means the "Letters" issue.

  ApJS = Astrophysical Journal Supplements Issue

  AJ = Astronomical Journal

  A&A = Astronomy & Astrophysics

  A&AS = Astronomy & Astrophysics Supplements

  PASJ = Publications of the Astronomical Society of Japan

Supplemental Sources:

Morras, R. & Mirabel, I.F. 1983, in The Milky Way Galaxy, IAU Symp. 106, eds. H. van Woerden, R.J. Allen, & W.B. Burton.

Weaver, W.F. 1974, in Galactic Radio Astronomy, IAU Symp. 60, eds. F.J. Kerr & S.C. Simonson III, 573.

Letter of 1994 Jan 27, Ken Croswell to R. Pogge.

Copyright © 1995, Richard W. Pogge, All Rights Reserved
1996 September 30 [rwp]