Vol.

UMz 12, NUMBER 18

PH

YSI CAI RK V I K%' I. KTTKRS

MESON OF MASS 959 MeV~

4 MAY 1964

OBSERVATION OF A NONSTRANGE

George R. Kalbfleisch, Luis W. Alvarez, Angela Barbaro-Galtieri, Orin I. Dahl, Philippe Eberhard, William E. Humphrey, James S. Lindsey, Deane%. Merrill, Joseph J. Murray, Alan Rittenberg, Ronald R. Ross, Janice B. Shafer, Frank T. Shively, Daniel M. Siegel, Gerald A. Smith, and Robert D. Tripp
Department
of Physics and Lawrence Radiation Laboratory, University (Received 9 April 1964)
of California,

Berkeley, California

%'e present here evidence showing the

exist-

'

'

959 MeV
Q

988eV60 0.86&1 (5~}&0.

ence of a nonstrange meson of mass 959 MeV. In the current experiment, the 72-in. hydrogen bubble chamber was exposed to a separated beam of 2. 45-, 2. 63-, and 2. 70-BeV/c K mesons from the Bevatron. Approximately 370 000 pictures were taken to date; approximately 300000 have been scanned. The reactions of interest in this paper are

g p, & s0.5BeV h

2
548 MeV

40-

{bj

lp

20~

'

+p-A

- A'+
-A

+MM,
7T++ m,

(1)
0.5
0.7
M

)i.ilB+
t. l
m+m

0::Fl ll
l. 5
l.T

0.9

I.3
m

l. g
M

0.2

0.3

0.4

0.5

(2)

(~

(BeV )

(~+~ ~ )

(BeV )

-A'+w++v'+m,
+n++7T +MM,

(3)
(4)

-A'+m++7T++m
A +7T +W

+m,
+7T

(5)
(6)

+'TT +'TT
m

Ao+

m+

+ m++

+ w + MM,

(7)

FIG. 1. (a) Distribution of the effective mass squared of ~+7T+7T m in the reaction@ +p A +71++7|++7t +7z The shaded area represents events in which the square of the momentum transfer from the proton to the lambda, is less than 0. 5 BeV2. (b) Distribution of the efQpA fective mass squared of all four ~+7I 7( combinations in the 35 A +7t +m +sr +7t +7to events in (a) where 0. 86 BeV M (57T} ~0. 98 BeV . The data are at incident momenta of 2. 45, 2. 63, and 2. 70 BeV/c.
m

3~++ 37T; A'+

meson.
where "MM*' stands for "missing mass. '* These reactions are found in the topologies of a V and 0, 2, 4, and 6 prongs. At this time we have measured the V-four-prong and ~ six-prong events in 250000 pictures, the V two-prongs in 135000 pictures, and the V zero-prongs in 100000 pic-

tures.
Figure 1(a) shows clearly the existence of the 959-MeV meson as an enhancement in the w+w+w w w spectrum from reaction (6). The mass is 959+ 2 MeV and the full width is l ~12 MeV. ' We observe 35 events in the interval 0. 86 ~ M (5w) ~ 0. 98 BeV . The background is estimated to be less than 10% of the peak. These events are produced mainly with a momentum ' less than 0. 5 BeV'. ' We have transfer g used the momentum transfer only as a means to separate other decay modes of this meson from the large background in reactions (1) through (5). The distribution of the four r+m'm combinations for each of the 35 events in the peak [Fig. 1(b)] clearly shows the presence of the 548-MeV q

Each of the 35 events has at least one triplet at the g mass. ' We conclude that the 959-MeV meson decays into m++ r +q. We now turn our attention to other possible decay modes of this meson. The neutral MM and the w+w MM distributions (at 2. 45 BeV/c only) from reactions (1) and (4) [Figs. 2(a) and 2(b)] show enhancements at 959 MeV. In addition, the selection of MM = g in the low-6' Aom+7T MM events gives a practically clean sample for the m+m g where the q decays into all neutrals [Fig. 2(b)]. This last selection gives a sample of 26 events with 0. 86 BeV'- M'(w+w MM) ~ 0. 98 BeV' and 0. 27 BeV' MM'~0. 33 BeV' for all the 135000 pictures in which these events were measured. ' No appreciable decay into TT+TT w+w w, and 2w"2w is observed [Figs. 2(c) and 2(d)]. In addition, no decay into 2w+2w 2w or 3m+37T is observed. ' We note that the decay rate into all neutrals [~20 A'+MM events above background in Fig. 2(a)] is comparable to the decay rate into w++w +q [four of the 35 A'+ 5w events occur in the 2. 45-BeV/c part of the sample in
527

VOX. U ME

12, NUMBER 1S

PHYSICAL REVIEW LETTERS

g +a++~ + MM &~p ~& 0.5 Bev~ 285 BeV/c
5 BeV&
O.ZS &

4 Mzv 1964

jP+

MM

Qg',
(w-p

& 0.5BeV' A 2.45 BeV/c

A+

events

MM'&0. 52BeV'

subtracted )

120-

959

(a)
959

MeV,

(b)

80-

MeV

80-

' . gh

0 ':-f. 2 0 04 0.8 !.
-

h, K

--'-'

-:

Ii

1.

0 04 0.8
M
Combined

1.

1.

MMa

(BeVe )

(w+w-MM)
A ~+or-+

(BeV')
Aa7

go~+-~o

2.45 BeY/c
C3

6p A~ 0.5

BeV

2.45 BeV /c 05Bev' G 6, &

959
MeV

80-

959
MeV

(d)

80IMlf

0 04 08
Ms (w+w'w-)

12

16 (BeV )

0 0.4 08
M

2 (Wwand w+w+w-w-)

1.

1.

(BeV

FIG. 2. (a) Distribution of the square of the mass of all neutrals (MM) in the reaction K +p A~+neutrals (MM). All events consistent with a fit to 7t +P-A +K have been subtracted from the plot. (b) Distribution of the effective mass squared of 7t++ 7t +neutrals (MM) in the reaction K +p-A + x + 7t +neutrals (MM). The crosshatched events have been selected to be consistent with q (0. 28 BeV ~MM 32 BeV ). (c) Distribution of the square of the effective mass of m atom in the reaction K +p A +7t +710+m . (d) Distribution of the square of the effective + + mass of x+x and 7t+7t+x 71 in the reactions K +p A +m++71 and A +7). +7t +x +7t, respectively. Shaded events are those with low momentum transfer from the proton to the lambda Q&A ~ 0. 5 BeV ). All data are 2. 45 BeV/e incident K momentum.

-0.

addition to the 14 A + r++ m +MM events in

Fig. 2(b)]. %e now look at the properties of the

m+n-q

de-

cay mode in an attempt to determine the quantum numbers of this meson. When we construct the Dalitz plot of M2(w+7)) versus M'(w q) and their projections [Fig. 3(a)] for the 61 A'+ w++w +q events (35 A + 5w and 26 A'+ w + w + MM), ' we observe no particular structure in the m g distributions. The distribution of points in the Dalitz plot is consistent with uniformity. The -20-MeV spread in the observed full width of the meson mass smears out the points appreciably

[envelopes labeled 950 and 970 MeV in Fig. 3(a)]. However, the distribution in M'(w+w outside the q) [Fig. 3(b)] is not appreciably affected by the resolution. The m+w distribution appears to be enhanced around 360 MeV. We now consider the mmg system as a di-pion and q, with no finalstate interaction between the q and the pions outside it, and denote the angular momentum of the di-pion as l and that between the di-pion and the states except 0+ are allowed. q as L. All J = 0 or 1, the possible spins and parities For are 0 (l=L=O); 1 (i =0, L= 1, or f= 1, L=O); and 1 (f= L= 1). For 0 and there can be no

1,

VOL U ME

12, NUMBER 18

PHYSICAL REVIEW LETTERS

M~

4 Mwv 1964

(~+~- outside qo) (BeV ) 0.2 O. I

{b)

0.7
e!I

Q Q

a =A~+w+wov
~

a (55events)
(26events)

=Am+~- hIM

20-

0
06-'

In summary, we have observed a meson of mass 959+2 MeV, '3 full width I'&12 MeV, and isospin T =0 or 1, which decays into m++m +q. No appreciable decay into two, three, four, or six pions is observed. The angular correlation of the g and the m+ in the di-pion rest system is not sin'8, If we asmaking unlikely the J+= 1 assignment.

0
(c)
gg Unotnblguous
h5~
S)na a

0 5 "''': -
.

O-

0,4;
0.4
{~+~o)

0.5
(BeY
)

0.6

0.7

-l.O -0.5

Cos

0 0.5 e~&

I.O

sume a strong decay (nonzero-width), the meson has isospin T =0 and G parity =+1, with 8& probably 0 or 1+. ' However, electromagnetic decay cannot be ruled out. The path length at 2. 45 BeVl c is approximately 0. 6 p, b per event. The -44 events Attest) at 2. 45 BeV/c yield a cross section of about 25 p. b. The average polarization of the A produced in association with this meson is

eA&=0+ 0. 2.
We wish to acknowledge

FIG. 3. (a) Dalitz plot [PI (m q ) versus M2(n q ) and projections for the 61 events in the reactions K +p-A +5m and M (5m) A +7(++x +neutrals with 0. 86 BeV or M2(m+7t + MM =go) 0. 98 BeV2. (b) Distribution of the square of the effective mass of the 7)+x outside the qo for the 61 events shown in (a). (c) Distribution of the cosine of the angle included between the qo and the as viewed in the di-pion (m m ) rest frame. Shaded Ao+7t++ z events are those from the reaction g P + neutrals {q~).

helpful discussions

with

Professor Murray Gell-Mann, Professor Donald H. Miller, and Dr. Nicola Cabibbo. We are indebted to the operators of the 72-in. bubble chamber and the Bevatron for their skill and patience. Furthermore, we thank our scanning and measuring staffs for their untiring efforts, without which this experiment would not have been possible.

correlation between the direction of the g and that of one of the pions as viewed in the di-pion rest

a sin'6)correlation is reframe. ' For 8 quired. The distribution in cosewn as observed in the di-pion rest frame is essentially isotropic, disagreeing with the sin ~&& prediction of a 1 state [Fig. 3(c)]. The ambiguity of the choice of the g in the A+ 5m events does not alter this conclusion. ' Thus, we conclude that the 959-MeV meson is probably not a vector particle. The isospin of the meson is either T=O or T=1 because of its production from an initial K P system in association with a A'. Absence of appreciable decay into m++m, m++m +m-, 27t++2m, 2n++2m + 2m, and 3m++3m implies the absence of appreciable decay into nm'. If the decay into m++7' +p occurs strongly, then the neutral mode m + m'+ g determines the isospin to be T = 0. How"zero" width for this meson is not exever, a The predicted branching cluded (1 & 12 MeV). ratios for a strongly decaying T =0 mmq system are in fair agreement with the observed ratios. In addition, observation of six e+e conversion pairs and four Compton electrons in association with the -50 A'+ MM events in the 959-MeV region points to a high multiplicity of gammas in the all-neutral decay, consistent with m + m'+ q. '

1,

"

"

~Work done under the auspices of the U. S. Atomic Energy Commission. 'The observed full width is -20 MeV, and the resolution is -16 MeV for the fitted A 7t. 7t n m m reaction. ~The square of the momentum transfer from the pro ton to the lambda is defined by &P A~ -{Ep -EA) + (PP PA) 2. Qf the 35 events with 0. 86 BeV~ Mt{5~) O. 98 BeV, twelve have one, eighteen have two, three have three, and two have four 7r+x g triplets near the g mass. The q mass is defined by 0. 282 BeV M~(z 7) m ) 0. 322 BeV~. This distribution is consistent with a random coincidence based on the phase space for the m~ decay [see Fig. 1(b)] and the resolution. 4The resolution for the MM and the m+7( MM is -25 MeV, which is essentially due to the 3 /& momentum spread in the beam. The observed full width in these MM channels of -30 MeV gives a full width for the 959 MeV meson of -20 NeV. ~Three Ao four-prong event s six-prong anti 28 A are consistent with A +6m or Z +5m in the total sample (-300000 scanned pictures). All 31 of these events M2(67t or y5~) 1. 1 BeV . 8In the A~+57l. events we chose as the q that ~+~ z triplet whose mass squared is closest to M&2=0. 300. This choice does not always pick the correct triplet because of the finite resolution. About half of the ambiguities should be chosen correctly statistically, giving 48 (of which 38 are unambiguous) correctly and 13 in-

Var. UMx 12, NUMsaR 18

PHYSICAL REVIEW LETTERS

body phase-space factor. This is in marked disagreement with the observed all-neutral to 7) m q rates. 'We use a branching ratio of -2. 5 for (q decay into all neutrals)/(charged decay) and 0. 5 for 2~0/7( 7( (7 =0). For the -44 events at low g at 2. 45 BeV/c, this predicts an apportionment of 8, 21, 4, and 11

correctly chosen (see reference 3). 7The Jp = 1 states can also be formed from (/=L =1 dr l=L =2), ($ =2, 5 =1 or l =1, J =2), and (l=2, L =2) combinations, respectively. The 0 (/ =L = 1) state would give a cos2g~ distribution, in disagreement with the data. The angular correlations would be

0, 1,

relatively complicated in these other cases. ~The odd intrinsic parity of the ~my system requires for a 1 matrix element (proportional to P x P) a N. Cabibbo points out that vector-meson assignment. the resulting sin20~ distribution is essentially independent of any m m interaction that might be present [such as at 360 MeV, Fig. 3(b)]. ~The A + 5' that are unambiguous as to the choice of the q are shown cross-hatched in Fig. 3(c). Replotting A + 5m so that all possible combinathe ambiguous tions contribute equally to the unit area per event does not appreciably alter the distribution presented in

events,
v

Fig. 3(c).
'OThe T = 0 component of the vector-meson octet has 7 4+~=01 . If the 71~ state with 0=+1 is this 01 The 1 meson, it must decay electromagnetically. vector assignment is unlikelybecause of the cos8~ distribution [Fig. 3(c)]. An additional argument can Charge-conjugation be made against this assignment. invariance in the electromagnetic decay requires a '=1 r 7I g system, so that all neutral decay must be 7. due to the modes z y, 2z y, and 3y. However, the decay rate into mop relative to 7r+7( q is expected to be very large. The &roy rate is proportional to e times a large two-body phase-space factor, whereas the TI+m p decay is proportional to n2 times a smaller three-

resPectively, as 7( 7( v)ch, m vr qneut, and all neutral. We observe 4, 14, ~6, and ~20, respectively, in fair agreement with the preIn addition, we note that we dicted apportionment. have not observed any v v~qo enhancement in Z (nv)+ states. This constitutes weak evidence against g =1. ' None of the 10 detected y events was consistent with A+y+MM for The MM in K +P Ayy production. these events is peaked from 500 to 900 MeV. This suggests an average of approximately 4 to 5 gammas for each of the 50 A + MM events. The observed detection efficiency for the gammas is then 10/(4 to 5) & 50=4 j(;, a reasonable value for the 72-in. chamber. ~3We note that evidence for an enhancement in the missing mass opposite the A in the reaction X +p-A+neutrals near this value has been observed by the Brookhaven bubble chamber group. [See M. Goldberg, M. Gundzik, J. Leitner, S. Lichtman, P. L. Connolly, E. L. Hart, K. %'. Lai, G. London, G. C. Moneti, R. R. Rau, N. P Samios, I. Q. Skillicorn, and S. S. Yamamoto, Bull. Am. Phys. Soc. 9, 23 (1964).] '4The existence of a singlet 00 + meson as well as 1 mesons has long been conjectured by M. Gell-Mann. J. Schwinger has also proposed a 0 i g meson at a ms of -1500 MeV [Phys. Rev. Letters 12, 237 (1964}].

zqchg,

SELF-CONSISTENCY OF SUPERGLOBAL MULTIPLET ASSIGNMENTS*

R. E.

Cutkosky
Pennsylvania

Carnegie Institute of Technology, Pittsburgh, (Received 2 April 1964)

irreducible representation in Schwinger's' SU(3) 8 SU(3) superglobal symmetry model may be denoted by (R, R, ), where R, and R, denote With respect to an "overSU(3) representations. all" SU(3), whose generators G= G, +Ga include the charge and hypercharge operators, this representation decomposes as R, (3 R2. The baryons are assumed to belong to (3'3), and the mesons may be taken as either (8 1) or (1 8). It has been suggested'y' that the & meson-baryon resonances be assigned to (3 8), which decomposes into (10) 0+(8). We shall show here that this assignment is incompatible with simple dynamical arguments. The baryon-exchange forces in (3 3) 8(1 8) are exactly the same as those which have been calculated~ in the Sakata model extra (3) in the the baryon designation do not enter into the couplings.
An

These forces therefore have the ratios given in s+ the last row of Table I, and in the ~ states are attractive in (3 H) but repulsive in (3 8). In order to obtain attraction in a (10) we have to start resonances with (8 1) mesons, which lead to in the representation (15 3) =(27) 0+(10) $(8). This seems like an excessive number of states, especially as the predicted resonances with Y = 2 or I=2 have not been found. It is instructive to compare this result with the global symmetry [SU(2) SSU(2)] model of I'=0 interactions, which leads to degenerate resonances with I=1 and I=2.' An independent argument in favor of assigning the mesons to (8 1) is obtained by comparing the pion-nucleon coupling strength to the other meson-baryon couplings. Note that the transformation (R, R, ) (R, R, ) is equivalent to Gell-Mann's
~~
~