The recently announced results by the CDF Collaboration, Study of multi-muon events produced in p-pbar collisions at sqrt(s)=1.96 TeV has been seen before, by the cosmic ray emulsion experiment JACEE. Since few particle physicists follow cosmic ray stuff, I thought I’d describe the JACEE results in more detail.
Some JACEE papers are given at the University of Washington’s Particle Astrophysics Group Webpage. See Excessive production of electron pairs in low multiplicity interactions or Observation of early photon conversions in high-energy cosmic-ray interactions. But my favorite arXiv paper for a long description of emulsion cosmic ray experimental results is Are Centauros exotic signals of the QGP?, Ewa Gladysz-Dziadus, 150pp, (2001).
A “Centauro” is an event with an excess of hadrons relative to electromagnetics. The opposite event, the “anti-Centauro” is an excess of electromagnetics (muons) over hadrons. (For non particle people, I’ll put an explanation of what “electromagnetics” has to do with “muons” below the fold.)
For the particle people, it’s best to just quote the paper. This is section “3.3 JACEE Experiment”, around page 63:
The Japanese-American Cooperative Emulsion Chamber Experiment, JACEE, has flown emulsion chambers with baloons near the top of the atmosphere. Despite of a small area and short time of exposure, as compared to Chacaltaya/Pamir Experiment, a few events of anomalous γ/charged ratio have been observed by JACEE Collaboration. However, these events differ in some essential points from classical Centauros. The anomalies were noticed at incident energies lower than that estimated for “classical” Centauros and unusual γ/charged ratios were observed only in the limited (η − φ) phase space region. Besides that, an excess of photons (anti–Centauro), in contrary to the hadron excess observed in Centauros, was claimed. The examples are:
1.) 4L-II-27 event  of incident energy of 80 TeV, yielded 149 charged particles and 120 γ’s. Almost all γ–quanta were produced in a narrow jet in the extreme forward direction. The γ/charged ratio is 2.6 ± 1.1 in the region of pseudorapidity 5.5 ≤ η ≤ 7.5, what is a significant deviation from the expected ratio of ∼ 1. The analysis presented in  showed an anomaly at the 5-10% level among 41 studied events with E0 ≥ 40 TeV.
2.) The event, with ΣE = 15.4 TeV, described in [75, 76] was initiated by a singly charged primary. The collision occurred within the detector. Almost all leading particles were γ–quanta. Photons appear to cluster into two groups. The leading cluster consisted of about 32 γ’s with hpT i ≃ 200 MeV and only one accompanying charged particle. A possibly distinct cluster had three times as many photons as charged hadrons (about 54 photons versus 17 charged). This event is one out of a sample of about 70.
3.) The event presented in  is a peripheral collision of Fe nucleus (E ≃ 9 TeV/nucleon) in emulsion. There were found 27 γ–quanta with η ≤ 6. As they came from pair conversions at only 0.8 radiation lengths, one can expect that the total number of photons was about 50. At the same time, only 6 charged particles (out of 21 charged tracks detected in the whole angular region) falled in the same kinematical range.
In all these events there was observed a tendency to a group emission of mesons. Such groups, having similar directions and momenta, could be signs of a formation and a subsequent decay of the chiral condensates. It should be mentioned, however, that these events were found in emulsion by scanning for the leading photon showers, so there was a “trigger bias” in favour of a large neutral fraction. It would be interesting to hear something about anti-Centauros from the mountain-top emulsion chambers. Here, there is, however, even much more stronger “trigger bias” in favour of gamma families, and thus the interpretation of data, from this point of view, is a complicated exercise. It is rather difficult to identify anti–Centauros unambigously, with exception of unusual and rare events in which the interaction vertex is close to the top and clearly resolved in the chamber.