hep-ph0302109/xi.tex
1: \documentstyle[12pt,epsf]{article}
2: %\pagestyle{empty}
3: \setlength{\oddsidemargin}{7.2pt}
4: \setlength{\topmargin}{0in}
5: \setlength{\textwidth}{6.2in}
6: \setlength{\textheight}{9in}
7: \setlength{\parskip}{0.2in}
8: \headheight0in
9: \headsep0in
10:  
11: \begin{document}
12:  
13: \newcommand{\be}{\begin{eqnarray}}
14: \newcommand{\ee}{\end{eqnarray}}
15: \begin{flushright}
16: LBNL-52160
17: \end{flushright}
18: \vspace{3mm} 
19: \begin{center}
20:  
21: {\Large
22: {$\Xi^-$ and $\Omega$ Distributions in Hadron-Nucleus Interactions\footnote{
23: This work was supported in part by the Director, Office of Science,
24: Division of Nuclear Physics of the Office of High Energy
25: and Nuclear Physics of the U. S. Department of Energy under Contract
26: Number DE-AC03-76SF00098.}
27: }} \\[8ex]
28: 
29: R. Vogt$^{a,b}$ and T.D. Gutierrez$^a$\\[2ex]
30:  $^a$Physics Department\\
31:  University of California at Davis\\
32:  Davis, California\quad 95616  \\
33:  and\\
34:  $^b$Nuclear Science Division\\
35:  Lawrence Berkeley National Laboratory\\
36:  Berkeley, California\quad 94720 \\[6ex]
37: 
38: {\bf ABSTRACT}
39: \end{center}
40: \begin{quote}
41: Strange baryons have long been known to exhibit a leading particle effect.
42: A recent comparison of $\Xi^-$ production in $\pi^-$, $n$, and $\Sigma^-$
43: interactions with nuclei show this effect clearly.  These data are 
44: supplemented by earlier 
45: measurements of $\Xi^-$ and $\Omega$ production by a $\Xi^-$ beam.  We
46: calculate the $\Xi^-$ and $\Omega$ $x_F$ distributions and nuclear dependence
47: in $hA$ interactions using the intrinsic model.
48: \end{quote}
49: \vspace{2mm}
50: \begin{center}
51: PACS numbers: 12.38.Lg, 13.85.Ni, 14.20.Jn
52: \end{center}
53: \newpage
54: 
55: \section{Introduction}
56: 
57: Leading particle effects, flavor correlations
58: between the final-state hadron and the projectile valence quarks, have 
59: long been observed in strange particle production.  Although many experiments
60: have recently focused on leading charm production 
61: \cite{E791,ddata1,ddata2,ddata3,ddata4,ddata5,wa891,wa892,SELEX1,SELEX2},
62: the first data involved strange particles 
63: \cite{bourq1,bourq2,bourq3,biagi,schn,beret}.  With new data from the WA89
64: collaboration on $\Xi^-$ production by $\pi^-$, $n$, and $\Sigma^-$ 
65: projectiles on nuclear targets \cite{wa89xi}, 
66: in addition to $\Xi^-$ production data from 
67: $\Xi^-$ beams \cite{biagi}, doubly strange hadron production can be studied
68: as a function of the number of strange valence quarks in the 
69: projectile.  We compare our model calculations to both the $x_F$
70: distributions and the integrated $A$ dependence reported by WA89 \cite{wa89xi}.
71: We also discuss $\Xi^-$ and $\Omega$ production by the $\Xi^-$ beam
72: \cite{biagi}. 
73: 
74: The WA89 collaboration used carbon, C, and copper, Cu, 
75: targets to study the $A$ 
76: dependence of $\Xi^- (dss)$ production by $\pi^-(\overline u d)$, $n(udd)$, 
77: and $\Sigma^- (dds)$ beams \cite{wa89xi}.  
78: The negative beams, $\pi^-$ and $\Sigma^-$, had
79: an average momentum of 345 GeV with a 9\% momentum spread.  
80: The neutron beam had a lower momentum with a larger
81: spread than the negative
82: beams---the average momentum was 260 GeV with a
83: 15\% variation.  The detected $\Xi^-$ was in the forward $x_F$ 
84: region, $x_F \geq 0.05$, with low transverse momentum, $p_T \leq 2.5$ GeV.
85: The data were parameterized in the form
86: \be
87: \frac{d\sigma}{dp_T^2 dx_F} \propto (1-x_F)^a e^{-bp_T^2} \, \, .
88: \label{param1}
89: \ee
90: The pion and neutron results agree with 
91: the functional form of Eq.~(\ref{param1}) over
92: all $x_F$.  For the pion, $a = 3.8 \pm 0.3$ for C and $4.1 \pm 0.3$ for Cu
93: while for the neutron $a = 5.0 \pm 0.3$ for C and $4.8 \pm 0.3$ for Cu.  These
94: results are consistent with expectations from spectator 
95: counting rules \cite{Gun}, $d\sigma/dx_F \propto (1-x_F)^{2n_s - 1}$.
96: With an incident gluon, $n_s = 2$ for pions and 3 for neutrons,
97: consistent with no leading particle effect for projectiles with zero 
98: strangeness.  There is no significant $A$ dependence of the exponent $a$.
99: 
100: On the other hand, the $\Sigma^-$ data cannot be fit to Eq.~(\ref{param1}) 
101: for $x_F < 0.4$.  In
102: the large $x_F$ region, $a = 2.08 \pm 0.04$ for C and $1.97 \pm 0.04$ for Cu.
103: These results indicate a very hard $x_F$ distribution, 
104: inconsistent with the counting rules even for a valence quark since $n_s = 2$ 
105: gives $(1-x_F)^3$.  In addition, at $x_F < 0.4$, the
106: distribution is independent of $x_F$ for both targets.  Thus
107: these data show a strong leading particle effect since the $\Xi^-$ has two 
108: valence quarks in common with the $\Sigma^-$.
109: The statistics are also sufficient for an observable $A$ dependence 
110: in the fitted values of $a$.
111: 
112: The integrated $A$ dependence was also reported by WA89
113: \cite{wa89xi}.  The $A$ dependence of the total cross section
114: is often parameterized as
115: \be
116: \sigma_{pA} = \sigma_{pp} A^\alpha \, \, .
117: \label{alfdef}
118: \ee
119: The integrated $\alpha$ for $\Sigma^-$ production of $\Xi^-$, 
120: $\alpha = 0.679 \pm 0.011$ \cite{wa89xi}, is 
121: in relatively good agreement with previous fits.  However, the pion
122: and neutron data show a closer-to-linear $A$ dependence, 
123: $\alpha = 0.891 \pm 0.034$ 
124: and $0.931 \pm 0.046$ respectively.  WA89 attributes this difference to
125: the fact that two $s \overline s$ pairs must be produced to make the 
126: final-state $\Xi^-$ and $s \overline s$ pair production would be suppressed 
127: relative to light $q \overline q$ production.  
128: 
129: WA89 has also measured the dependence of $\alpha$ on $x_F$. 
130: This dependence, $\alpha(x_F)$, was previously reported for a wide range of 
131: hadron projectiles \cite{Geist}.  For non-strange hadrons and hadrons with a
132: single strange quark, there is a common trend with $x_F$.
133: At $x_F = 0$, $\alpha \approx 0.8$ and decreases to
134: $\approx 0.5$ at large $x_F$, an overall decrease of $\sim A^{1/3}$ for
135: $0 < x_F < 1$.  The $\Xi^0$, the only doubly-strange hadron included in
136: Ref.~\cite{Geist}, is an exception.  In
137: $pA$ interactions, the $\Xi^0$ has a larger value of $\alpha$ at low $x_F$
138: \cite{beret}.  A similar effect is observed for $\Xi^-$ production
139: by WA89.  Their measurements of $\alpha(x_F)$ for $\Xi^-$ from 
140: pion and neutron beams show that $\alpha \sim 1$ for $x_F
141: \sim 0.05$, decreasing to $\alpha \sim 0.7$ at higher $x_F$.  
142: Thus the decrease of $\alpha$ with $x_F$ is also $A^{1/3}$ in this case
143: although the actual values of $\alpha$ are larger than those for lighter
144: hadrons \cite{Geist}.  However, for $\Sigma^-$-induced $\Xi^-$ production, 
145: $\alpha \sim 0.7$ almost independent of $x_F$.  
146: 
147: The other data we consider are $\Xi^- \, {\rm Be} \rightarrow \Xi^-,
148: \Omega(sss)$ at 116 GeV, measured by Biagi {\it et al.}\ \cite{biagi}.  
149: In this case, the final-state $\Xi^-$ $x_F$ distribution 
150: increases with $x_F$, as does the $\Omega$ $x_F$ 
151: distribution.  This increase could be due in part to
152: the use of an invariant parameterization\footnotetext{For the two
153: parameterizations to be equivalent, the right-hand side of Eq.~(\ref{param1})
154: should be multiplied by $2E/\sqrt{s}$ to obtain the invariant cross section.}
155: \cite{biagi},
156: \be
157: E \frac{d\sigma}{dp^3} \propto (1-x_F)^{a'} e^{-b' p_T^2} \, \, ,
158: \label{param2}
159: \ee
160: which fits the $\Xi^-$ data at $x_F > 0.5$ but only 
161: approximately fits the $\Omega$ data in this limited region.
162: The exponent $a'$ was fit in two $p_T^2$ intervals, $p_T^2 < 0.4$ GeV$^2$ and
163: $0.4 < p_T^2 < 2.9$ GeV$^2$, yielding $a' = -0.45 \pm 0.02$ and 
164: $-0.18 \pm 0.03$ respectively.  Between the most central measurement, 
165: $x_F = 0.15$, and the 
166: projectile fragmentation region, $x_F = 0.85$, the $\Xi^-$ cross section 
167: increases by a factor of $\sim 40$ in the low $p_T^2$ interval.  
168: 
169: A comparison of these results with incident proton data \cite{biagi}, 
170: $pA \rightarrow \Xi^- X$ \cite{bourq1,Cardello}
171: and $pA \rightarrow pX$ \cite{Barton},
172: showed that, at low $x_F$, the $\Xi^-$ production cross section
173: is essentially independent of the projectile while, at high $x_F$, the 
174: $\Xi^-$ and $p$ scattering cross sections are similar.  This behavior
175: supports valence quark domination at high $x_F$.  The structure of the $\Omega$
176: $x_F$ distribution is similar:  it is of the same order of magnitude as $pA
177: \rightarrow \Omega X$ \cite{bourq1} at low $x_F$ 
178: but is similar to singly strange baryon production 
179: by protons, $pA \rightarrow \Lambda X$ and $\Sigma^+ X$, 
180: \cite{bourq1,Cardello,Skubic} at high $x_F$.
181: 
182: Since only one target was used, $\alpha = 0.6$ was assumed in
183: Eq.~(\ref{alfdef}) to obtain 
184: the per nucleon cross sections.  This extrapolated
185: cross section is a factor of $1.5-2$ higher than those on hydrogen
186: targets \cite{biagi}.  An extrapolation with $\alpha = 1$ 
187: gives better agreement with the hydrogen target data, at least for
188: $\Xi^-$ production.
189: 
190: We employ the intrinsic model \cite{intc1,intc2,VB,VBlam,GutVogt1}
191: as developed for strangeness production in Ref.~\cite{GutVogtIS}. 
192: In the intrinsic model, a hadron can fluctuate into Fock
193: state configurations with a combination of light and strange quark pairs. 
194: The heavier quarks in the configuration are comoving with the other
195: partons in the Fock state and thus can coalesce with these comoving partons
196: to produce strange hadrons at large $x_F$. 
197: The model combines leading-twist production of $s \overline s$ pairs with
198: intrinsic Fock states with up to nine particles.  Thus coalescence production
199: of the $\Omega$ from a proton is possible.  
200: 
201: \section{Leading-Twist Production}
202: 
203: We calculate leading-twist strangeness in perturbative QCD, assuming the
204: strange quark is massive.  When the projectile has nonzero strangeness, we also
205: consider the possibility of flavor excitation.  We choose proton parton
206: distribution functions with the lowest possible initial scale $\mu_0^2$ so that
207: $m_s^2 > \mu_0^2$.  Therefore the baryon parton distribution functions are
208: based on the GRV 94 LO proton parton distributions \cite{GRV94}.
209: We use the most recent pion parton densities by Gl\"{u}ck, Reya
210: and Schienbein \cite{GRSpi2}.  
211: To be conservative, we assume that the scale $\mu$ at which the strong coupling
212: constant and the parton densities are evaluated is $\mu = 2m_T$ where $m_T = 
213: \sqrt{p_T^2 + m_s^2}$ and $m_s = 500$ MeV.
214: The $x_F$ distributions, obtained by integrating Eq.~(\ref{param1}) or
215: (\ref{param2}) over $p_T$, are dominated by low $p_T$ production.
216: 
217: We treat the strange quark as heavy, as in Ref.~\cite{GutVogtIS}, 
218: rather than as a massless parton in jet-like 
219: processes.  Treating the strange quark as a jet means 
220: selecting a minimum $p_T$ to keep the cross section finite.  A large minimum
221: $p_T$ compatible with hard scattering is incompatible with the
222: assumption of intrinsic production, inherently a low $p_T$
223: process \cite{VB}.  Strange hadrons can either be produced directly in jet
224: production or by the fragmentation of light quark
225: and gluon jets.  However, there is no indication that these data originate 
226: from jets.
227: 
228: The leading-twist $x_F$ distribution of heavy quark production \cite{VBH2} 
229: is denoted by $F$,
230: \be
231: F \equiv
232: \frac{d\sigma_{\rm ltf}^S}{dx_F} = 
233: \frac{\sqrt{s}}{2} \int dz_3\, dy_2\, dp_T^2  x_a x_b H_{AB}(x_a,x_b,\mu^2)
234: \frac{1}{E_1}\ \frac{D_{S/s}(z_3)}{z_3}\ \, \, ,
235: \label{ltfus}
236: \ee
237: where $A$ and $B$ are the initial hadrons, $a$ and $b$ 
238: are the interacting partons, 1 and 2 are the 
239: produced strange quarks and 3 is the final-state strange hadron $S$.
240: The $x_F$ of the detected quark is $x_F = 2m_T \sinh y/\sqrt{s}$ where $y$
241: is the rapidity of the quark and $\sqrt{s}$ is the hadron-hadron center of mass
242: energy.  We assume the simplest possible fragmentation function,
243: \be D_{S/s}(z) = B_S \delta(1-z) \,\, , \label{fusfrag} \ee
244: with $B_S = 0.1$, assuming that all
245: 10 ground-state strange hadrons are 
246: produced at the same rate to leading twist \cite{GutVogtIS}.  This choice
247: of $D_{S/s}$ gives the hardest possible leading twist $x_F$ distribution.
248: 
249: The convolution of the subprocess $q \overline q$
250: annihilation and gluon fusion cross sections 
251: with the parton densities is included in $H_{AB} (x_a, x_b, \mu^2)$,
252: \be
253: H_{AB}(x_a,x_b,\mu^2) & = & \sum_q [f_q^A(x_a,\mu^2) 
254: f_{\overline q}^B(x_b,\mu^2) + f_{\overline q}^A(x_a,\mu^2) f_q^B(x_b,\mu^2)] 
255: \frac{d \widehat{\sigma}_{q \overline q}}{d \hat{t}} \label{hab} \\
256:  &   & \mbox{} + f_g^A(x_a,\mu^2) f_g^B(x_b,\mu^2) \frac{d
257: \widehat{\sigma}_{gg}}{d \hat{t}} \, \, , \nonumber
258: \ee
259: where $q = u,$ $d$, and
260: $s$.  Although including the $s$ quark in the sum over $q$ in Eq.~(\ref{hab}) 
261: could lead to some over counting,
262: the strange quark contribution to $F$ from non-strange projectiles is 
263: negligible, less than $0.01$\% for neutron and pion projectiles.  It is
264: somewhat larger for strange
265: projectiles, 2.5\% for the $\Sigma^-$ and 5.6\% for the $\Xi^-$ but it is
266: only significant at large $x_F$.  
267: 
268: Hyperon parton distributions can be
269: inferred from the proton distributions \cite{GutVogt1} by simple counting
270: rules.  We can relate the valence $s$ distribution of the $\Sigma^-$,
271: $f_{s_v}^{\Sigma^-}$, to the proton valence $d$ distribution, $f_{d_v}^p$,
272: and the valence $d$ distribution in the $\Sigma^-$,
273: $f_{d_v}^{\Sigma^-}$, 
274: to the valence $u$ in the proton, $f_{u_v}^p$, so that
275: \begin{eqnarray}
276: \int_0^1 dx \, f_{s_v}^{\Sigma^-} (x,\mu^2) & = & \int_0^1 dx \, f_{d_v}^p 
277: (x,\mu^2) = 1 \, \, , \\
278: \int_0^1 dx \, f_{d_v}^{\Sigma^-} (x,\mu^2) & = & \int_0^1 dx \, f_{u_v}^p 
279: (x,\mu^2) = 2 \, \, .
280: \label{partsumsig}
281: \end{eqnarray}
282: We also identify the up quark in the 
283: sea of the $\Sigma^-$ with the strange quark in the proton sea, 
284: $f_u^{\Sigma^-}(x,\mu^2) = f_s^p(x,\mu^2)$.  
285: Similar relations hold for the antiquark distributions.  Likewise,
286: for the $\Xi^-$, we relate the valence $s$, $f_{s_v}^{\Xi^-}$, to the 
287: valence $u$ in the proton, $f_{u_v}^p$, and
288: equate the valence $d$ distributions in both baryons so that,
289: \begin{eqnarray}
290: \int_0^1 dx \, f_{s_v}^{\Xi^-} (x,\mu^2) & = & \int_0^1 dx \, f_{u_v}^p 
291: (x,\mu^2) = 2 \, \, , \\
292: \int_0^1 dx \, f_{d_v}^{\Xi^-} (x,\mu^2) & = & \int_0^1 dx \, f_{d_v}^p 
293: (x,\mu^2) = 1 \, \, .
294: \label{partsumxi}
295: \end{eqnarray}
296: Here also, $f_u^{\Xi^-}(x,\mu^2) = f_s^p(x,\mu^2)$.  
297: The gluon distributions are assumed to
298: be identical for all baryons, $f_g^p = f_g^{\Sigma^-} = f_g^{\Xi^-}$.  
299: The leading order
300: subprocess cross sections for heavy quark production
301: can be found in Ref.~\cite{Ellis}.  
302: The fractional momenta carried by the projectile and target partons, 
303: $x_a$ and $x_b$,
304: are $x_a = (m_T/\sqrt{s}) (e^y + e^{y_2})$ and $x_b = (m_T/\sqrt{s}) (e^{-y} 
305: + e^{-y_2})$.
306: 
307: We have assumed only $gg$ and $q \overline q$ contributions
308: to massive strange quark production.
309: We have also checked how the $x_F$ distribution would
310: change if the strange quark was treated as massless and all $2 \rightarrow 2$
311: scattering channels were included.  
312: Jet production of $s$ quarks is through processes such as
313: $g s \rightarrow g s$, $q s \rightarrow qs$ and $\overline q s \rightarrow 
314: \overline q s$.  (Similarly for the $\overline s$.)
315: Including these processes increases the cross section by a factor 
316: of $4-8$.  While this factor
317: is not constant, it increases rather slowly with $x_F$ so that the difference
318: in shape is only important in the region where intrinsic production dominates,
319: as discussed later.
320: 
321: Contributions from massless $2 \rightarrow 2$ scattering 
322: increase more rapidly at $x_F >0$ for strange projectiles
323: because the contribution from, for example, $f_s^{\Sigma^-}(x_a) f_g^p(x_b)$, 
324: dominates the scattering cross section.  In the infinite momentum frame,
325: $f_{s_v}^{\Sigma^-} = f_{d_v}^p$, see Eq.~(\ref{partsumsig}), and
326: $f_{s_v}^{\Sigma^-}$ is large at large $x_a$ while $f_g^p$ increases as $x_b$ 
327: decreases. To take this into account quantitatively, 
328: we have incorporated ``flavor excitation'' of massive strange
329: valence quarks.  The excitation matrix elements for massive quarks
330: are found in Ref.~\cite{comb}.  The
331: flavor excitation cross section has a pole when $p_T \rightarrow 0$ so that
332: a cutoff, $p_{T_{\rm min}}$, is required to keep this cross
333: section finite, as in jet production.  We employ $p_{T_{\rm min}}
334: = 2m_s = 1$ GeV.  The leading-twist fusion cross section for strange
335: projectiles is then augmented by 
336: \be
337: X_{p_{T_{\rm min}}} \equiv
338: \frac{d\sigma_{\rm lte}^S}{dx_F} = 
339: \frac{\sqrt{s}}{2} \int  dz_3\, dy_2\, dp_T^2 
340: x_a' x_b' H_{AB}'(x_a',x_b',\mu^2)
341: \frac{1}{E_1}\ \frac{D_{S/s}(z_3)}{z_3}\ \, \, 
342: \label{ltexc}
343: \ee
344: where
345: \begin{eqnarray}
346: H_{AB}'(x_a',x_b',\mu^2) =  f_{s_v}^A(x_a',\mu^2) \{ \sum_q 
347: [f_q^B(x_b',\mu^2) + f_{\overline q}^B(x_b',\mu^2)] \frac{d
348: \widehat{\sigma}_{sq}}{d \hat{t}} + f_g^B(x_b',\mu^2) \frac{d
349: \widehat{\sigma}_{sg}}{d \hat{t}} \} \, \, , \label{habexc}
350: \end{eqnarray}
351: $x_a' = (m_Te^y + p_Te^{y_2})/\sqrt{s}$ and $x_b' = (m_Te^{-y} 
352: + p_Te^{-y_2})/\sqrt{s}$.  Note that there is no overlap between the processes
353: included in Eqs.~(\ref{habexc}) and (\ref{hab}) and thus no double
354: counting. This excitation mechanism is effective
355: only for hadrons with a strange quark in the final state and thus does not
356: affect the distributions with a produced $\overline s$.
357: 
358: To summarize, for strange and antistrange final states produced by non-strange 
359: hadrons,
360: \be \frac{d\sigma^S_{\rm lt}}{dx_F} =  \frac{d\sigma^S_{\rm ltf}}{dx_F} \equiv
361: F \, \, ,
362: \label{fdef}
363: \ee
364: as in Eq.~(\ref{ltfus}).  This relation also holds for antistrange final states
365: from strange hadrons.  However, for strange hadron production by 
366: hadrons with nonzero strangeness, we also consider
367: \be \frac{d\sigma^S_{\rm lt}}{dx_F} = \frac{d\sigma^S_{\rm ltf}}{dx_F} 
368: + \frac{d\sigma^S_{\rm lte}}{dx_F} \equiv
369: F + X_{p_{T_{\rm min}}} \, \, 
370: \label{efdef}
371: \ee
372: where flavor excitation, Eq.~(\ref{ltexc}), may play a role.
373: 
374: We remark that the role of flavor excitation in heavy quark production, as
375: outlined in Ref.~\cite{comb}, is questionable.  It was first proposed as a
376: leading order contribution to the total cross section and, as such, could be
377: rather large if the heavy quark distribution in the proton is significant.
378: However, the proton heavy quark distribution is only nonzero above the
379: threshold $m_Q$.  In addition, parton distribution functions are defined in the
380: infinite momentum frame where the partons are treated as massless.  Later
381: studies at next-to-leading order (NLO) \cite{NDE} showed that these excitation
382: diagrams are a subset of the NLO cross section and suppressed relative to
383: fusion production.  They are only a small fraction of the heavy flavor
384: cross section and thus play no significant role at low energies.  
385: Strange hadron production at large $x_F$ is then an important test of the
386: excitation process.  
387: 
388: \begin{figure}[htpb]
389: \setlength{\epsfxsize=0.95\textwidth}
390: \setlength{\epsfysize=0.5\textheight}
391: \centerline{\epsffile{xif1_grs.eps}}
392: \caption[]{Leading-twist strange quark production in
393: (a) $\pi^- p$ interactions at 345 GeV, (b) $np$ interactions at 260 GeV,
394: (c) $\Sigma^- p$ interactions at 345 GeV, and (d) $\Xi^- p$ interactions at
395: 116 GeV.  The solid curves are the fusion results, $F$.  For projectiles with
396: valence strange quarks, the excitation contributions, $X_{p_{T_{\rm min}}}$ 
397: with $p_{T_{\rm min}} = 1$ GeV, are shown in
398: the dashed curves. The dot-dashed curves are the total, $F + 
399: X_{p_{T_{\rm min}}}$.}
400: \label{fus}
401: \end{figure}
402: 
403: The leading-twist $x_F$ distributions with all four projectiles are shown in
404: Fig.~\ref{fus}.  We compare the fusion $x_F$ distributions, $F$, 
405: in $\pi^- p$ and $\Sigma^- p$ interactions at 345 GeV, $np$ interactions at 
406: 260 GeV, and $\Xi^- p$ interactions at 116 GeV, corresponding to the energies 
407: we investigate.  The $\pi^- p$ distribution is the broadest because
408: the $f_{\overline u_v}^{\pi^-}(x_a) f_{u_v}^p(x_b)$ 
409: contribution hardens the $x_F$
410: distribution at large $x_F$ where the $q \overline q$ channel dominates.  
411: When excitation is considered, as in $\Sigma^- p$ and $\Xi^- p$
412: interactions, the $x_F$ distribution is hardened, particularly through the
413: $f_s^{\Sigma^-}(x_a') f_g^p(x_b')$ and $f_s^{\Xi^-}(x_a') f_g^p(x_b')$
414: contributions.  These dominate at large $x_F$ where $f_s^S(x_a')$
415: is large for valence strange quarks and $f_g^p(x_b')$ is large at small
416: $x_b'$.  The effect is even stronger for the $\Xi^-$ since it 
417: has two valence strange quarks.   Note that,
418: except at small $x_F$, the total leading-twist cross section is equivalent to
419: $X_{p_{T_{\rm min}}}$.
420: 
421: We can obtain an approximate estimate of the exponent $a$ from
422: Eq.~(\ref{param1}) from the average $x_F$, $\langle x_F \rangle$,
423: where 
424: \be
425: a = \frac{1}{\langle x_F \rangle} - 2 \, \, .
426: \label{avea}
427: \ee
428: When averaged over $x_F>0$, the values of $a$ obtained are larger than
429: those measured by WA89, as expected.  For the pion and neutron beams, $a = 5.2$
430: and 8.7 respectively.  Strangeness production by strange hadrons including
431: fusion alone also gives large values of $a$, 9 for the $\Sigma^-$ and 7.4 for
432: the $\Xi^-$.  The $x_F$ distribution of strange quarks produced by flavor
433: excitation is considerably harder, $a = 3.2$ for the $\Sigma^-$ and 2.2 for
434: the $\Xi^-$.  Combining the two contributions, as in the dot-dashed curves
435: in Fig.~\ref{fus}(c) and (d), gives a somewhat larger value of $a$ than for
436: flavor excitation alone, $a = 4.7$ and 2.9 for $\Sigma^-$ and $\Xi^-$ beams
437: respectively.  The values of $a$ obtained from Eq.~(\ref{avea}) are all much
438: higher than those obtained from the data.  Thus the leading twist results alone
439: cannot explain the shape of the measured $\Xi^-$ $x_F$ distributions.
440: 
441: \section{The Intrinsic Model for Strangeness}
442: 
443: We now briefly discuss the intrinsic model for strangeness production, 
444: described in detail for $\pi^- p$ interactions in Ref.~\cite{GutVogtIS}.  
445: Since all the data is at $x_F > 0$, 
446: we only discuss intrinsic production in the projectile.
447: 
448: The hadron wavefunction is a superposition of Fock state fluctuations
449: in which the hadron contains one or more ``intrinsic'' $Q \overline Q$ pairs.
450: These pairs can hadronize when the hadron interacts, breaking the
451: coherence of the state.  The model, first developed for charm 
452: \cite{intc1,intc2}, gives heavy quarks a
453: larger fraction of the projectile momentum due to their greater mass.  
454: The strange quark is lighter so that the momentum gained is not as large.
455: However, the intrinsic strangeness probability is larger, 
456: $P_{\rm is}^5 \sim 2$\%.  For simplicity, we assume that the intrinsic 
457: probabilities are independent of the valence quark content of the projectile.
458: Then $P^5_{\rm is}$ is identical for nucleons and hyperons.  The Fock state
459: probabilities for up to 3$Q \overline Q$ pairs where at least one $Q \overline
460: Q$ pair is strange are given in
461: Ref.~\cite{GutVogtIS}.  
462: 
463: In this paper, we focus on $\Xi^-(dss)$ and $\Omega(sss)$ production from
464: $\pi^- (\overline u d)$, $n(udd)$, $\Sigma^-(dds)$ and $\Xi^-(dss)$
465: projectiles.  The produced $\Xi^-$ shares one or more valence
466: quarks with the projectile.  We study $\Omega$ production only by $\Xi^-$
467: projectiles, with two valence quarks in common.
468: 
469: Once the coherence of the Fock state is broken, the partons in the state can
470: hadronize in two ways.  The first, uncorrelated fragmentation of the strange
471: quark, is the same basic mechanism as at leading twist.  However, when the Fock
472: state fluctuation includes all the valence quarks of the final-state hadron,
473: these quarks, in close spatial proximity, can coalesce into the final-state
474: hadron and come on shell.  Thus, to calculate the full strange and 
475: antistrange hadron $x_F$ distributions in
476: the intrinsic model, we include uncorrelated fragmentation of the strange
477: quark in every state considered and coalescence from those states where it is 
478: possible. In Ref.~\cite{GutVogtIS}, a comparison of the model with strange 
479: baryon asymmetries suggested that fragmentation may not be an effective
480: mechanism because when the Fock state has minimal invariant mass, 
481: fragmentation may cost too much energy.  This conclusion
482: needs to be checked against inclusive $x_F$ distributions over a broader $x_F$
483: range.    
484: 
485: In principle, the parton distributions of the hadron can be defined through
486: such a Fock-state expansion \cite{hoyerroy}.  In each fluctuation, only the
487: mass distinguishes the light and heavy quark distributions.  Thus it is not
488: really possible in a given state to separate the ``valence'' and ``sea''
489: distributions.  All are similar as long as the quarks are light. One
490: distinguishing feature is our assumption that only strange quarks can produce
491: strange final-state hadrons by uncorrelated fragmentation.
492: Thus with hyperon projectiles, uncorrelated
493: fragmentation may also be possible from Fock
494: states with only light $Q \overline Q$ pairs.  These intrinsic 
495: light quark states must be included in the total probability,
496: as described in Ref.~\cite{GutVogtIS}.  The probabilities
497: for these states must also be defined.  We assume
498: \be 
499: P^5_{\rm iq} & = & \left( \frac{\widehat{m}_s}{\widehat{m}_q} \right)^2 
500: P^5_{\rm is} \approx 5\% \, \, , \label{piq} \\
501: P^7_{\rm iqq} & = & \left( \frac{\widehat{m}_s}{\widehat{m}_q} \right)^2
502: P^7_{\rm isq} = 1.75 \, P^5_{\rm is} \, \, , \label{piqq} \\
503: P^9_{\rm iqqq} & = & \left( \frac{\widehat{m}_s}{\widehat{m}_q} \right)^4
504: P^9_{\rm issq} = 1.25 \, P^5_{\rm is} \, \, . \label{piqqq} 
505: \ee
506: We further assume that the probabilities for the meson Fock 
507: configurations are equal to the baryon probabilities.
508: 
509: We have only taken the 10 strange ground state hadrons and 
510: antihadrons into account.  
511: We assume that each hadron has a 10\% production probability 
512: from fragmentation, neglecting the particle masses. The final-state $x_F$
513: distribution is then equivalent to that of the $s$ or $\overline s$ quark. 
514: For coalescence, we count the number of 
515: possible strange and antistrange hadron combinations that can be obtained 
516: from a given Fock state and
517: assign each strange hadron or antihadron that fraction of the total.
518: The possible number of strange hadrons is greater than
519: the number of possible strange antihadrons.
520: We clearly err in the overall normalization by simply 
521: including the ground state strange particles.  However,
522: the higher-lying resonances have the same quark
523: content with the same fragmentation
524: and coalescence distributions since all properties of
525: the final-state hadrons except their quark content are neglected.
526: 
527: To obtain the total probability of each strange hadron to be produced from
528: projectile hadron, $h$, in the
529: intrinsic model, we sum 
530: the probabilities over all the states with up to 3 intrinsic $Q \overline Q$
531: pairs.  Thus
532: \be \frac{dP_S^h}{dx_F} = \sum_n \sum_{m_u} \sum_{m_d} \sum_{m_s} \beta
533: \left( \frac{1}{10} \frac{dP_{{\rm i} (m_s {\rm s}) (m_u {\rm u}) (m_d {\rm d})
534: }^{nF}}{dx_F} + \xi \frac{dP_{{\rm i} (m_s {\rm s}) (m_u {\rm u}) (m_d {\rm d})
535: }^{nC}}{dx_F} \right) \, \, .
536: \label{intsum}
537: \ee
538: To conserve probability, $\beta = 1$ when the hadron is only 
539: produced by uncorrelated fragmentation and 0.5 when both fragmentation and 
540: coalescence are possible.  When we assume coalescence production only, we 
541: set $P^{nF} \equiv 0$ and $\beta \equiv 1$.
542: The weight of each state produced by coalescence is 
543: $\xi$ where $\xi = 0$ when $S$ is not produced by coalescence in state
544: $|n_v  m_s(s \overline s) m_u(u \overline u) m_d(d \overline d) \rangle$.  
545: The number of up, down and strange $Q \overline Q$ pairs 
546: is indicated by $m_u$, $m_d$ and $m_s$ respectively.  The total, $m_u + m_d
547: + m_s = m$, is defined as $m = (n - n_v)/2$ because each $Q$ in an $n$-parton
548: state is accompanied by a $\overline Q$. For baryon projectiles, $n =5$, 7, 
549: and 9 while for mesons $n=4$, 6, and 8.  Depending on the value of $n$, $m_i$
550: can be 0, 1, 2 or 3, {\it e.g.}\ in a $|uud s \overline s d \overline d d 
551: \overline d \rangle$ state, $m_u = 0$, $m_d = 2$ and $m_s = 1$ with $m = 3$.  
552: Note that $m_s=0$ is only possible when $h$ is strange since
553: no additional $s \overline s$ pairs are thus needed to produce some strange
554: hadrons by coalescence.
555: The total probability distributions, $dP_S^h/dx_F$, for $S = \Xi^-$ and 
556: $\Omega$ are given in the Appendix.
557: 
558: \section{$A$ Dependence of Combined Model}
559: 
560: The total $x_F$ distribution for final-state strange hadron $S$ is the sum
561: of the leading-twist fusion and intrinsic strangeness components,
562: \be
563: \frac{d\sigma^S_{hN}}{dx_F} = \frac{d\sigma^S_{\rm lt}}{dx_F} +
564: \frac{d\sigma^S_{\rm iQ}}{dx_F} \, \, .
565: \label{ismodel}
566: \ee
567: The leading-twist distributions are defined in Eqs.~(\ref{fdef}) and 
568: (\ref{efdef}).  The total intrinsic cross section, $d\sigma^S_{\rm iQ}/dx_F$, 
569: is related to $dP_S^h/dx_F$ by
570: \be
571: \frac{d\sigma^S_{\rm iQ}}{dx_F} = \sigma_{h N}^{\rm in}
572: \frac{\mu^2}{4 \widehat{m}_s^2} \frac{dP_S^h}{dx_F} \, \, .
573: \label{iscross}
574: \ee
575: 
576: We assume that the relative $A$
577: dependence for leading-twist and intrinsic production is the same as that for
578: heavy quarks and quarkonia \cite{VB,GutVogt1,VBH2,VBH1}.  The $A$ dependence of
579: the two component model is
580: \be
581: \sigma_{hA} = A^\beta \sigma_{\rm lt} + A^\gamma \sigma_{\rm iQ}
582: \label{sigadep}
583: \ee
584: where the combination of the two terms should approximate $A^\alpha$
585: in Eq.~(\ref{alfdef}).
586: There are no strong nuclear effects on open charm at leading twist so that the 
587: $A$ dependence is linear at $x_F \sim 0$, $\alpha = 1$, \cite{e789d}, 
588: dropping to $\alpha = 0.77$ for 
589: pions and 0.71 for protons \cite{VBH2,Badier} at higher $x_F$ where the
590: intrinsic model begins to dominate. We assume that $\beta = 1$ and $\gamma = 
591: 0.77$ for pions and 0.71 for all baryons.  Thus, 
592: \be
593: A^{\gamma - \beta} \approx A^{-1/3} \,\,\,\, {\rm as} \,\,\,\,
594: x_F \rightarrow 1 \, \,
595: . \label{tcadep}
596: \ee
597: This relative $A$ dependence, similar to that discussed earlier for 
598: light hadrons \cite{Geist}, is
599: included in our calculations.  
600: The proton and neutron numbers are taken into account in the calculation of
601: the leading-twist cross section.  This isospin effect is small for fusion, $F$,
602: which is dominated by the $gg$ channel.  In perturbative QCD, $\beta = 1$ 
603: could be modified by changes in the nuclear parton distributions relative to
604: the proton \cite{Arneodo}.  However, the scale for our perturbative
605: calculation is too low for such models of
606: these modifications to apply \cite{EKS981,EKS982} and are not
607: included in our calculations.
608: 
609: \section{Results}
610: 
611: We begin by comparing the model to the WA89 pion data in Fig.~\ref{pipa}.
612: These data do not strongly distinguish between leading-twist
613: fusion and the full model.  The intrinsic results do
614: not significantly depend on uncorrelated
615: fragmentation.  All three curves agree rather well with the data, primarily
616: because the fusion $x_F$ distribution is already fairly hard.  Then the
617: intrinsic contribution is a small effect even though
618: the $d$ valence quark is common between the $\pi^-$ and the $\Xi^-$.  This is
619: perhaps due to the fact that $P_{\rm iss}^6$ is already rather small, $\sim
620: 0.6$\%. We note that the calculated total cross sections agree with the
621: measured cross sections to within better than a factor of two despite the
622: rather large uncertainties in the calculations.
623: \begin{figure}[htpb]
624: \setlength{\epsfxsize=0.95\textwidth}
625: \setlength{\epsfysize=0.3\textheight}
626: \centerline{\epsffile{xif2_grs.eps}}
627: \caption[]{The model is compared to the 345 GeV WA89 pion data on
628: (a) C and (b)
629: Cu targets.  The dotted curves are leading-twist fusion, $F$, alone, the dashed
630: curves include uncorrelated fragmentation and coalescence, and the solid
631: curves include coalescence alone.  The data sets have been normalized to 
632: the cross section per nucleon.  The curves are normalized to the data at $x_F =
633: 0.15$.}
634: \label{pipa}
635: \end{figure}
636: 
637: Even though the intrinsic contribution is relatively small, it significantly
638: affects the value of $a$ obtained from Eq.~(\ref{avea}).  The difference
639: between the $a$ values found without and with uncorrelated fragmentation in the
640: intrinsic model is
641: negligible for the pion beam.  We find $a \approx 4.1$ for the C target
642: and 4.3 for the Cu target relative to $a=5.2$ for leading twist alone.  
643: These results are within the errors of the WA89 fit
644: to their data.  The agreement is especially good  
645: since the two-component model does not give a smooth falloff as a
646: function of $x_F$ that can be easily quantified by a single exponent.
647: 
648: It is also possible to calculate $\alpha(x_F)$ and the $x_F$-integrated
649: $\alpha$ from the ratios of the distributions.  The calculations including
650: both uncorrelated fragmentation and coalescence generally give a smaller value
651: of $\alpha$ and, hence, a stronger $A$ dependence.  This is because
652: fragmentation peaks at low $x_F$, influencing the $A$ dependence sooner than
653: coalescence alone which is only significant at intermediate $x_F$.
654: Thus $\alpha(x_F) \sim 0.9$ for fragmentation and coalescence while $\alpha$
655: decreases from $\sim 1$ at low $x_F$ to 0.86 at high $x_F$ with coalescence
656: alone.  The integrated values are 0.93 and 0.98 respectively, somewhat higher
657: than the WA89 result but with the same general trend.
658: 
659: The overall agreement with the total cross section is not as good for the $nA$
660: data, shown in Fig.~\ref{npa}.  Surprisingly, the distribution
661: including uncorrelated fragmentation agrees best with
662: the data.  This is perhaps because the energy of the secondary
663: neutron beam is least well determined.  The
664: energy spread is 15\% compared to 9\% for the pion and $\Sigma^-$ beams.  
665: A small energy variation can have a large effect on the 
666: leading-twist
667: cross section.  A 15\% increase in the average neutron energy, from 260 GeV to 
668: 300 GeV, increases $d\sigma_{\rm lt}/dx_F$ by 80\% at $x_F \sim 0.25$ while the
669: intrinsic cross section is essentially unaffected.  Such a shift in the
670: relative leading-twist and intrinsic production rates could easily reduce the
671: effect of coalescence alone to be more compatible with the data.
672: The uncertainty in the energy of the pion beam has a much weaker effect on the
673: final result because the intrinsic contribution is already small, as is obvious
674: from Fig.~\ref{pipa}.
675: \begin{figure}[htpb]
676: \setlength{\epsfxsize=0.95\textwidth}
677: \setlength{\epsfysize=0.3\textheight}
678: \centerline{\epsffile{xif3.eps}}
679: \caption[]{The model is compared to the 260 GeV WA89 neutron data on
680: (a) C and (b)
681: Cu targets.  The dotted curves are leading-twist fusion, $F$, alone, the dashed
682: curves include uncorrelated fragmentation and coalescence, and the solid
683: curves include coalescence alone.  The data sets have been normalized to
684: the cross section per nucleon. The curves are normalized to the data at $x_F =
685: 0.15$.} 
686: \label{npa}
687: \end{figure}
688: 
689: The calculated exponents $a$ are larger for the neutron than the pion, in
690: agreement with the WA89 measurements \cite{wa89xi}. We find $a \approx 4.9$
691: for C and 5.8 for Cu.  Typically the value of $a$ obtained for fragmentation
692: and coalescence is larger than that for coalescence alone since eliminating the
693: fragmentation contribution tends to increase $\langle x_F \rangle$.  The
694: stronger $A$ dependence assumed for the intrinsic model has the effect of
695: increasing $a$ for larger nuclei.  Thus the $a$ found for the carbon target
696: agrees rather well with the WA89 data while the copper data suggest a harder
697: distribution than our calculation implies.  There is, however, a stronger 
698: $A$ dependence in the falloff with $x_F$ than in the data which cannot
699: distinguish between the values of $a$ determined for the two targets.  This
700: stronger dependence is reflected in the values of $\alpha$ obtained, 0.87 when
701: fragmentation and coalescence are included and 0.95 with coalescence alone.
702: 
703: Even though there is some $A$ dependence in the model calculations,
704: the relatively small intrinsic contribution to the pion and neutron data
705: leads to a rather weak overall $A$ dependence.  Dominance of the leading-twist
706: cross section at low to intermediate $x_F$ results in a nearer-to-linear 
707: integrated $\alpha$, as observed by WA89 \cite{wa89xi}.
708: 
709: We now turn to $\Xi^-$ production by the $\Sigma^-$ where the $A$ dependence
710: can be expected to be stronger.  For the
711: first time, we have a valence strange quark in the projectile so that we can
712: compare the effectiveness of fusion alone with flavor
713: excitation.  We can also better test the importance of uncorrelated
714: fragmentation because coalescence production is already possible
715: in the 5-parton Fock state, $|dds s \overline s
716: \rangle$. 
717: 
718: Our results are collected in Fig.~\ref{sigmpa}.  
719: We first discuss the
720: importance of uncorrelated fragmentation to leading-twist
721: fusion, $F$, alone, Fig.~\ref{sigmpa}(a) and (b).  
722: The leading-twist contribution is rather steeply falling.
723: Including both uncorrelated fragmentation and
724: coalescence broadens the $x_F$ distribution but cannot match the hardness of
725: the measured $x_F$ distribution.  Eliminating the fragmentation contribution
726: produces a much harder distribution for $x_F \geq 0.15$, matching the
727: shape of the data relatively well.
728: \begin{figure}[htpb]
729: \setlength{\epsfxsize=0.95\textwidth}
730: \setlength{\epsfysize=0.5\textheight}
731: \centerline{\epsffile{xif4.eps}}
732: \caption[]{The model is compared to the 345 GeV WA89 $\Sigma^- A$ data on C and
733: Cu targets.  In (a) and (b), the leading twist contribution is $F$ 
734: while in (c) and (d), flavor excitation is also included, $F +
735: X_{p_{T_{\rm min}}}$. 
736: The dotted curves are for leading-twist alone, the dashed
737: curves include uncorrelated fragmentation and coalescence and the solid
738: curves include coalescence alone.  The data sets have been normalized to
739: the cross section per nucleon. The curves are normalized to the data at $x_F =
740: 0.15$.}
741: \label{sigmpa}
742: \end{figure}
743: 
744: The calculations are all normalized to the $x_F = 0.15$ point to better
745: compare the shapes of the distributions. Fragmentation gives better agreement 
746: at low $x_F$ because uncorrelated
747: fragmentation peaks close to $x_F \sim 0$, filling
748: in the low to intermediate $x_F$ range.  Coalescence, on the other hand, always
749: produces strange hadrons with $\langle x_F \rangle \geq 0.3$, broadening the
750: distribution only in this region.  Thus without 
751: fragmentation the
752: calculations overestimate the
753: data at $x_F \sim 0$.  The data seem to indicate that
754: uncorrelated fragmentation is not an effective mechanism for intrinsic
755: production, in agreement with the conclusions of Ref.~\cite{GutVogtIS}.
756: 
757: The agreement with the solid curves in Fig.~\ref{sigmpa}(a) and (b)
758: is good but not perfect.  The calculation overestimates the data at high $x_F$.
759: Recall that for the neutron, the 15\% spread in the beam momentum
760: could result in an overestimate of the intrinsic contribution, as previously
761: discussed.  Although the possible spread in the $\Sigma^-$ beam momentum
762: is smaller, it could affect the relative intrinsic 
763: contribution at low to intermediate 
764: $x_F$.  At large $x_F$, the effect on the shape would be negligible because the
765: intrinsic contribution dominates.  Thus, given the inherent uncertainties
766: in the model and in the data, the agreement is rather satisfactory.
767: 
768: We have obtained the value of the exponent $a$ from $\langle x_F \rangle$,
769: both over all $x_F$ and for $x_F > 0.1$, avoiding the strong change in slope 
770: of the solid curves when coalescence alone is included in the intrinsic result.
771: When the entire forward $x_F$ range is integrated over, $a = 3.02$ for C and
772: 3.39 for Cu with both uncorrelated fragmentation and coalescence while with
773: coalescence alone, $a = 1.24$ for C and 1.78 for Cu.  Considering only the
774: range $x_F > 0.1$, we find $a = 1.56$ for C and 1.63 for Cu with fragmentation
775: and 0.43 for C, 0.57 for Cu without fragmentation.  The calculated $a$'s
776: suggest considerably harder $x_F$ distributions in the more limited $x_F$
777: region, particularly when coalescence alone is considered.  However, none of 
778: the results are in good agreement with $a \approx 2$, as obtained by WA89 for 
779: $x_F \geq 0.4$.  This is not surprising, especially since the solid curve is
780: seen to be harder than the data for $x_F > 0.1$.  In any case, the coalescence
781: contributions in particular, now considerably more important than for $\Xi^-$
782: production by non-strange hadrons, are difficult to fit to a power law since
783: they approach zero at both $x_F = 0$ and $x_F = 1$ with a peak at intermediate
784: $x_F$, see the curves in Ref.~\cite{GutVogtIS}.  The various contributions,
785: all with somewhat different magnitudes due to the relative probabilities, peak
786: at different values of $x_F$, complicating the situation further.
787: 
788: The calculated values of $\alpha$ give $\alpha \approx 0.8$ for the integrated
789: cross sections but $\alpha \approx 0.7$ for $x_F > 0.4$, with and without
790: fragmentation in the intrinsic model, rather consistent with the WA89 result.
791: However, as a function of $x_F$, coalescence alone is more consistent with the
792: measurements since $\alpha \approx 1$ at $x_F \approx 0$, decreasing to 0.71
793: as $x_F \rightarrow 1$, as expected from Eq.~(\ref{sigadep}). 
794: 
795: We now check if our results improve when we include 
796: flavor excitation, Eq.~(\ref{efdef}), shown in
797: Fig.~\ref{sigmpa}(c) and (d).  Now the
798: baseline leading twist distribution, $F + X_{p_{T_{\rm min}}}$, is harder than
799: with fusion alone, as shown in Fig.~\ref{fus}(c).  
800: However, although the distribution
801: is broader, it still drops six orders of magnitude over the entire $x_F$ range
802: with $p_{T_{\rm min}} = 1$ GeV.  Thus including
803: flavor excitation cannot describe the data without the intrinsic coalescence
804: component, as in
805: Fig.~\ref{sigmpa}(a) and (b). The total cross section is in reasonable
806: agreement with that measured by WA89. Decreasing $p_{T_{\rm min}}$ further
807: can harden the distribution but still underestimates the data.
808: A lower $p_{T_{\rm min}}$ enhances the total cross sections considerably 
809: so that, if $p_{T_{\rm min}} = 0.25$ GeV, the cross section is overestimated by
810: several orders of magnitude.  The
811: intrinsic contribution is then negligible so that decreasing $p_{T_{\rm min}}$
812: actually degrades the agreement with the data.  Thus there is no clear evidence
813: for flavor excitation.
814: 
815: The trends in the $A$ dependence are similar when excitation is included
816: although the values of $a$ obtained are suggestive of harder $x_F$
817: distributions than with leading-twist fusion alone.  In particular, the
818: excitation contribution is harder at low $x_F$, see Fig.~\ref{fus}, causing
819: the change in slope due to the hardening of the intrinsic distribution when
820: coalescence alone is included to be less abrupt.  Nonetheless, the agreement
821: with the measured value of $a$ is not significantly improved.  The calculated
822: values of $\alpha(x_F)$ are similar to those with leading-twist fusion alone
823: but the integrated values of $\alpha$ are somewhat larger, $\approx 0.87$,
824: due to the larger leading-twist baseline.  The $A$ dependence also does not
825: support flavor excitation as a significant contribution to strange hadron
826: production. 
827: 
828: To summarize, the $A$ dependence of $\Xi^-$ production by $\Sigma^-$
829: is stronger because the intrinsic contribution
830: with coalescence dominates the $x_F$ distribution already at $x_F \sim
831: 0.1$.  Therefore the integrated $A$ dependence is nearly a factor of
832: $A^{1/3}$ down relative to the pion and neutron data, as shown in
833: Eqs.~(\ref{sigadep}) and (\ref{tcadep}).  Thus the trends of the model are
834: in agreement with the WA89 data.
835: 
836: Finally, we compare our intrinsic model calculations with the invariant $\Xi^-$
837: and $\Omega$ cross sections measured in $\Xi^-$Be interactions at 116 GeV
838: \cite{biagi}.
839: Since intrinsic production is expected to be a primarily low $p_T$ effect
840: \cite{VB}, we only compare to the low $p_T^2$ bin, $0<p_T^2 < 0.4$
841: GeV$^2$ \cite{xiomdat}.  The data and our calculations are 
842: shown in Fig.~\ref{ximpa}.  We have multiplied our
843: $x_F$ distributions by $2m_T \cosh y/\sqrt{s}$ to obtain the invariant cross
844: section.  The invariant $x_F$ distributions are harder as a function of
845: $x_F$. 
846: \begin{figure}[htpb]
847: \setlength{\epsfxsize=0.95\textwidth}
848: \setlength{\epsfysize=0.5\textheight}
849: \centerline{\epsffile{xif5.eps}}
850: \caption[]{The model is compared to the 116 GeV $\Xi^-$Be data.  
851: In (a) and (b), the leading twist contribution is $F$ 
852: while in (c) and (d), flavor excitation is also included, $F +
853: X_{p_{T_{\rm min}}}$. 
854: The dotted curves are for leading-twist fusion alone, the dashed
855: curves include uncorrelated fragmentation and coalescence, the solid
856: curves include coalescence alone and the dot-dashed curves include a
857: diffractive ``Pomeron'' contribution.  The data sets have been normalized to
858: the cross section per nucleon. The curves are normalized to the data at $x_F =
859: 0.15$.}  
860: \label{ximpa}
861: \end{figure}
862: 
863: Because the initial and final states are identical for $\Xi^-$ production, the
864: intrinsic contribution increases with $x_F$.  However, even with coalescence
865: alone, the increase does not continue beyond $x_F \sim 0.4$.  A similar but
866: weaker effect is seen for the $\Omega$ where there are two strange quarks in
867: common with the $\Xi^-$.  Therefore we have tried to identify
868: a mechanism that would increase the cross section
869: beyond $x_F \sim 0.4$.  One possibility is a
870: ``Pomeron-like'' parton in the Fock state.  Since the Pomeron has
871: quantum numbers similar to two gluons, it can be exchanged between two 
872: projectile valence quarks.  A $|dss {\cal P} \rangle$ state, where ${\cal P}$
873: signifies the ``Pomeron'', would result in $\Xi^-$ states at high $x_F$ while
874: avoiding the $\delta(1-x_F)$ delta function for the 3-particle Fock state.
875: A $\Xi^-$ from such a configuration would have a distribution peaking at $x_F
876: \rightarrow 1$.  It is also possible to imagine a $|dss s
877: \overline s {\cal P} \rangle$ state from which both
878: the $\Xi^-$ and $\Omega$ could
879: be produced.  In this case, the distribution would peak away from $x_F \sim 1$.
880: We included ``Pomeron'' production from both states assuming that $P_{{\rm
881: i}{\cal P}}^4 = P_{\rm iq}^5 \sim 5$\% and that  $P_{{\rm
882: is}{\cal P}}^6 = P_{\rm iqq}^7 \sim 3.5$\%, giving these configurations large
883: probability.  The results, shown in the dot-dashed curves in Fig.~\ref{ximpa},
884: agree relatively well with the data, especially for the $\Xi^-$.
885: The $\Omega$ data are still underestimated but the trend is now in the right
886: direction. 
887: 
888: Of course, this is a rather artificial solution, especially when 
889: the initial and final states are not identical.  If it is correct, it
890: should also be included in $\Sigma^- p
891: \rightarrow \Xi^- X$, as shown in Fig.~\ref{sigmpa}.  However, we have checked
892: this and found that the resulting $x_F$ distribution is far too
893: hard.  Therefore, the practicality of the mechanism is questionable and the
894: ``Pomeron'' results should not be taken too seriously.
895: 
896: Our calculations with flavor
897: excitation are compared to the data in Fig.~\ref{ximpa}(c) and (d).
898: The results do not improve, even when the ``Pomeron'' is
899: included.  Indeed, the results with excitation underestimate the data at 
900: high $x_F$ more than with fusion alone.
901: Therefore we conclude that flavor excitation is
902: not an effective mechanism for strange hadron production at low $p_T$,
903: in keeping with the interpretation of the excitation
904: diagrams as NLO contributions to the production cross section, 
905: as discussed previously.
906: 
907: We have also calculated the exponent $a'$, see Eq.~(\ref{param2}), for these
908: distributions with $x_F > 0.5$.  The results are negative for all the cases
909: shown. We find $a' \approx -0.45$ with and without uncorrelated fragmentation
910: and $-0.5$ with the ``Pomeron''.  These values are rather consistent with
911: those obtained for the low $p_T^2$ selection of the $\Xi^-$ data.  The
912: corresponding values for $\Omega$ production are somewhat lower, $\approx
913: -0.41$, without the ``Pomeron'' but somewhat higher, $\approx -0.54$, with it.
914: 
915: \section{Conclusions}
916: 
917: We have compared our intrinsic calculations to $\Xi^-$
918: production by $\pi^-$, $n$ and $\Sigma^-$ projectiles and
919: to $\Xi^-$ and $\Omega$ production by $\Xi^-$ projectiles.  We find good
920: agreement with the WA89 data for leading-twist fusion
921: and coalescence.  Flavor excitation 
922: seems excluded as a significant mechanism of low $p_T$ strange
923: hadron production.  The apparent difficulties with uncorrelated fragmentation 
924: seen in Ref.~\cite{GutVogtIS} are confirmed here.  Thus
925: coalescence production is the most effective mechanism for strange hadron 
926: production in the intrinsic model.
927: The leading charm analysis should perhaps be revisited in light of 
928: this conclusion.
929: 
930: The conclusions that can be reached from the $\Xi^-$-induced interactions 
931: at 116 GeV are less clear.  It is possible that a ``Pomeron-like'' 
932: state could exist in the
933: hadron wavefunction but its applicability to $\Omega$ production is
934: somewhat doubtful.  Therefore the interpretation of
935: these data within the
936: intrinsic model is rather inconclusive.  More standard studies of diffractive
937: production in $\Xi^- \, {\rm Be} \, \rightarrow \Xi^- X$ should be performed.
938: 
939: Acknowledgments: We thank
940: P. Hoyer for reminding us of these data.  R.V. would like to thank
941: the Niels Bohr Institute and the Grand Accelerateur National d'Ions Lourds
942: for hospitality at the beginning of this work.
943: 
944: 
945: %\newpage
946: \setcounter{equation}{0}
947: \renewcommand{\theequation}{\arabic{equation}}
948: \begin{center}
949: {\bf Appendix}
950: \end{center}
951: \vspace{0.2in}
952: 
953: Here we give the relevant probability distributions in the intrinsic model
954: for $\Xi^-$ and $\Omega$
955: production used in our calculations.  To more clearly distinguish between the
956: probability distributions including uncorrelated fragmentation and
957: coalescence and those with coalescence alone, both distributions are given.
958: 
959: First, we give the distributions relevant to the WA89 measurements.
960: We reproduce the $\Xi^-$ probability distribution from a $\pi^-$
961: \cite{GutVogtIS},
962: \be
963: \frac{dP_{\Xi^-}^{\pi^-}}{dx_F} & = & \frac{1}{10} \frac{dP_{\rm
964: is}^{4F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
965: isu}^{6F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
966: isd}^{6F}}{dx_F} + 
967:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
968: iss}^{6F}}{dx_F} + \frac{1}{7} \frac{dP_{\rm iss}^{6C}}{dx_F} \right) \nonumber
969: \\ &   & + \, \frac{1}{10} \frac{dP_{\rm
970: isuu}^{8F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
971: isud}^{8F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
972: isdd}^{8F}}{dx_F} + 
973:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
974: issu}^{8F}}{dx_F} + \frac{1}{12} \frac{dP_{\rm issu}^{8C}}{dx_F} \right) 
975: \nonumber \\ &   & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
976: issd}^{8F}}{dx_F} + \frac{2}{12} \frac{dP_{\rm issd}^{8C}}{dx_F} \right) +
977:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
978: isss}^{8F}}{dx_F} + \frac{3}{16} \frac{dP_{\rm isss}^{8C}}{dx_F} \right)\,\, , 
979: \label{piprobxim} \\
980: %\ee
981: %\be
982: \frac{dP_{\Xi^-}^{\pi^-}}{dx_F} & = & \frac{1}{7} 
983: \frac{dP_{\rm iss}^{6C}}{dx_F} + \frac{1}{12} 
984: \frac{dP_{\rm issu}^{8C}}{dx_F} + \frac{2}{12} \frac{dP_{\rm issd}^{8C}}{dx_F}
985: + \frac{3}{16} \frac{dP_{\rm isss}^{8C}}{dx_F} \, \, .
986: \label{piprobximco} 
987: \ee
988: From a neutron projectile,
989: \be
990: \frac{dP_{\Xi^-}^n}{dx_F} & = & \frac{1}{10} \frac{dP_{\rm
991: is}^{5F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
992: isu}^{7F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
993: isd}^{7F}}{dx_F} + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
994: iss}^{7F}}{dx_F} + \frac{1}{13} 
995: \frac{dP_{\rm iss}^{7C}}{dx_F} \right) \nonumber
996: \\ &   & + \, \frac{1}{10} \frac{dP_{\rm
997: isuu}^{9F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
998: isud}^{9F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
999: isdd}^{9F}}{dx_F} +
1000:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1001: issu}^{9F}}{dx_F} + \frac{2}{22} \frac{dP_{\rm issu}^{9C}}{dx_F} \right) 
1002: \nonumber \\ &   & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1003: issd}^{9F}}{dx_F} + \frac{3}{22} \frac{dP_{\rm issd}^{9C}}{dx_F} \right) +
1004:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1005: isss}^{9F}}{dx_F} + \frac{6}{28} \frac{dP_{\rm isss}^{9C}}{dx_F} \right)\,\, , 
1006: \label{nprobxim} \\
1007: \frac{dP_{\Xi^-}^n}{dx_F} & = & \frac{1}{13} 
1008: \frac{dP_{\rm iss}^{7C}}{dx_F} + \frac{2}{22} \frac{dP_{\rm issu}^{9C}}{dx_F} 
1009: + \frac{3}{22} \frac{dP_{\rm issd}^{9C}}{dx_F} + \frac{6}{28} 
1010: \frac{dP_{\rm isss}^{9C}}{dx_F} \, \, . 
1011: \label{nprobximco} 
1012: \ee
1013: The $\Xi^-$ distribution from a $\Sigma^-$ projectile is,
1014: \be
1015: \frac{dP_{\Xi^-}^{\Sigma^-}}{dx_F} & = & \frac{1}{10} \frac{dP_{\rm
1016: iu}^{5F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
1017: id}^{5F}}{dx_F} + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1018: is}^{5F}}{dx_F} + \frac{2}{6} \frac{dP_{\rm is}^{5C}}{dx_F} \right) + 
1019: \frac{1}{10} \frac{dP_{\rm iuu}^{7F}}{dx_F} 
1020: + \frac{1}{10} \frac{dP_{\rm iud}^{7F}}{dx_F}  
1021: \nonumber \\ &  & + \,
1022: \frac{1}{10} \frac{dP_{\rm idd}^{7F}}{dx_F} 
1023: + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1024: isu}^{7F}}{dx_F} + \frac{1}{13} 
1025: \frac{dP_{\rm isu}^{7C}}{dx_F} \right) 
1026: + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1027: isd}^{7F}}{dx_F} + \frac{3}{13} \frac{dP_{\rm isd}^{7C}}{dx_F} \right) 
1028: \nonumber \\ &   & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1029: iss}^{7F}}{dx_F} 
1030: + \frac{6}{16} \frac{dP_{\rm iss}^{7C}}{dx_F} \right) 
1031: %\nonumber \\ &   & + \, 
1032: + \frac{1}{10} \frac{dP_{\rm
1033: iuuu}^{9F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm iuud}^{9F}}{dx_F} 
1034: + \frac{1}{10} \frac{dP_{\rm
1035: iudd}^{9F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
1036: iddd}^{9F}}{dx_F} \nonumber \\ &   & + \, 
1037: \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1038: isuu}^{9F}}{dx_F} + \frac{2}{22} \frac{dP_{\rm isuu}^{9C}}{dx_F} \right) 
1039: + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1040: isud}^{9F}}{dx_F} 
1041: + \frac{3}{22} \frac{dP_{\rm isud}^{9C}}{dx_F} \right)
1042: \nonumber \\ &   & + \, 
1043: \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1044: isdd}^{9F}}{dx_F} + \frac{14}{22} \frac{dP_{\rm isdd}^{9C}}{dx_F} \right) +
1045:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1046: issu}^{9F}}{dx_F} + \frac{6}{28} \frac{dP_{\rm issu}^{9C}}{dx_F} \right) 
1047: \nonumber \\ &   & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1048: issd}^{9F}}{dx_F} + \frac{9}{28} \frac{dP_{\rm issd}^{9C}}{dx_F} \right) +
1049:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1050: isss}^{9F}}{dx_F} + \frac{12}{32} \frac{dP_{\rm isss}^{9C}}{dx_F} 
1051: \right)\,\, , 
1052: \label{sigprobxim} \\
1053: \frac{dP_{\Xi^-}^{\Sigma^-}}{dx_F} & = & \frac{2}{6} 
1054: \frac{dP_{\rm is}^{5C}}{dx_F} + \frac{1}{13} 
1055: \frac{dP_{\rm isu}^{7C}}{dx_F} + \frac{3}{13} \frac{dP_{\rm isd}^{7C}}{dx_F} 
1056: + \frac{6}{16} \frac{dP_{\rm iss}^{7C}}{dx_F} + \frac{2}{22} 
1057: \frac{dP_{\rm isuu}^{9C}}{dx_F} 
1058: \nonumber \\ &  & + \, \frac{3}{22} \frac{dP_{\rm isud}^{9C}}{dx_F}
1059: + \frac{14}{22} \frac{dP_{\rm isdd}^{9C}}{dx_F} + \frac{6}{28} 
1060: \frac{dP_{\rm issu}^{9C}}{dx_F} + \frac{9}{28} \frac{dP_{\rm issd}^{9C}}{dx_F}
1061: + \frac{12}{32} \frac{dP_{\rm isss}^{9C}}{dx_F} \, \, .
1062: \label{sigprobximco} 
1063: \ee
1064: 
1065: We now present the relevant probability distributions for $\Xi^-$ and $\Omega$
1066: production from a $\Xi^-$ projectile.  First we give the $\Xi^-$ distributions,
1067: \be 
1068: \frac{dP_{\Xi^-}^{\Xi^-}}{dx_F} & = & 
1069: \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1070: iu}^{5F}}{dx_F} + \frac{1}{6} \frac{dP_{\rm iu}^{5C}}{dx_F} \right) 
1071: + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1072: id}^{5F}}{dx_F} + \frac{2}{6} \frac{dP_{\rm id}^{5C}}{dx_F} \right) \nonumber
1073: \\ &  & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1074: is}^{5F}}{dx_F} + \frac{3}{7} \frac{dP_{\rm is}^{5C}}{dx_F} \right) + 
1075:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1076: iuu}^{7F}}{dx_F} + \frac{1}{13} \frac{dP_{\rm iuu}^{7C}}{dx_F} \right) 
1077: \nonumber
1078: \\ &   & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1079: iud}^{7F}}{dx_F} + \frac{2}{13} \frac{dP_{\rm iud}^{7C}}{dx_F} \right) +
1080:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1081: idd}^{7F}}{dx_F} + \frac{3}{13} \frac{dP_{\rm idd}^{7C}}{dx_F} \right) 
1082: \nonumber \\ &  & + \,
1083:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1084: isu}^{7F}}{dx_F} + \frac{3}{16} 
1085: \frac{dP_{\rm isu}^{7C}}{dx_F} \right) +
1086: \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1087: isd}^{7F}}{dx_F} + \frac{6}{16} \frac{dP_{\rm isd}^{7C}}{dx_F} \right)
1088: \nonumber \\ &   & + \, 
1089:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1090: iss}^{7F}}{dx_F} 
1091: + \frac{6}{17} \frac{dP_{\rm iss}^{7C}}{dx_F} \right) +
1092:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1093: iuuu}^{9F}}{dx_F} + \frac{1}{22} \frac{dP_{\rm iuuu}^{9C}}{dx_F} \right)
1094: \nonumber \\ &   & + \,
1095:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1096: iuud}^{9F}}{dx_F} 
1097: + \frac{2}{22} \frac{dP_{\rm iuud}^{9C}}{dx_F} \right) +
1098:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1099: iudd}^{9F}}{dx_F} + \frac{3}{22} \frac{dP_{\rm iudd}^{9C}}{dx_F} \right)
1100: \nonumber \\ &   & + \, 
1101:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1102: iddd}^{9F}}{dx_F} 
1103: + \frac{4}{22} \frac{dP_{\rm iddd}^{9C}}{dx_F} \right) +
1104:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1105: isuu}^{9F}}{dx_F} + \frac{3}{28} \frac{dP_{\rm isuu}^{9C}}{dx_F} \right)
1106: \nonumber \\ &   & + \, 
1107:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1108: isud}^{9F}}{dx_F} 
1109: + \frac{6}{28} \frac{dP_{\rm isud}^{9C}}{dx_F} \right) +
1110:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1111: isdd}^{9F}}{dx_F} + \frac{9}{28} \frac{dP_{\rm isdd}^{9C}}{dx_F} \right)
1112: \nonumber \\ &   & + \, 
1113:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1114: issu}^{9F}}{dx_F} + \frac{6}{31} \frac{dP_{\rm issu}^{9C}}{dx_F} \right) +
1115:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1116: issd}^{9F}}{dx_F} + \frac{12}{31} \frac{dP_{\rm issd}^{9C}}{dx_F} \right)
1117: \nonumber \\ &   & + \, 
1118:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1119: isss}^{9F}}{dx_F} + \frac{12}{37} 
1120: \frac{dP_{\rm isss}^{9C}}{dx_F} \right)\,\, , \\
1121: \label{xiprobxi} 
1122: \frac{dP_{\Xi^-}^{\rm \Xi^-}}{dx_F} & = & 
1123: \frac{1}{6} \frac{dP_{\rm iu}^{5C}}{dx_F} 
1124: + \frac{2}{6} \frac{dP_{\rm id}^{5C}}{dx_F} + \frac{3}{7} 
1125: \frac{dP_{\rm is}^{5C}}{dx_F} + \frac{1}{13} \frac{dP_{\rm iuu}^{7C}}{dx_F} 
1126: + \frac{2}{13} \frac{dP_{\rm iud}^{7C}}{dx_F} + \frac{3}{13} 
1127: \frac{dP_{\rm idd}^{7C}}{dx_F} 
1128: \nonumber \\ &  & + \,
1129: \frac{3}{16} 
1130: \frac{dP_{\rm isu}^{7C}}{dx_F} + \frac{6}{16} \frac{dP_{\rm isd}^{7C}}{dx_F} 
1131: + \frac{6}{17} \frac{dP_{\rm iss}^{7C}}{dx_F} 
1132: + \frac{1}{22} \frac{dP_{\rm iuuu}^{9C}}{dx_F} 
1133: + \frac{2}{22} \frac{dP_{\rm iuud}^{9C}}{dx_F} 
1134: + \frac{3}{22} \frac{dP_{\rm iudd}^{9C}}{dx_F} \nonumber
1135: \\ &   & + \, 
1136: \frac{4}{22} \frac{dP_{\rm iddd}^{9C}}{dx_F} 
1137: + \frac{3}{28} \frac{dP_{\rm isuu}^{9C}}{dx_F} 
1138: + \frac{6}{28} \frac{dP_{\rm isud}^{9C}}{dx_F} 
1139: + \frac{9}{28} \frac{dP_{\rm isdd}^{9C}}{dx_F} \nonumber \\ &   & + \,
1140: \frac{6}{31} \frac{dP_{\rm issu}^{9C}}{dx_F} 
1141: + \frac{12}{31} \frac{dP_{\rm issd}^{9C}}{dx_F} + \frac{12}{37} 
1142: \frac{dP_{\rm isss}^{9C}}{dx_F} \,\, .
1143: \label{xiprobxico} 
1144: \ee
1145: Finally, we give the $\Omega$ distribution from a $\Xi^-$ projectile,
1146: \be
1147: \frac{dP_{\Omega}^{\rm \Xi^-}}{dx_F} & = & \frac{1}{10} \frac{dP_{\rm
1148: iu}^{5F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
1149: id}^{5F}}{dx_F} + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1150: is}^{5F}}{dx_F} + \frac{1}{7} \frac{dP_{\rm is}^{5C}}{dx_F} \right) + 
1151: \frac{1}{10} \frac{dP_{\rm iuu}^{7F}}{dx_F} 
1152: + \frac{1}{10} \frac{dP_{\rm iud}^{7F}}{dx_F}  
1153: \nonumber \\ &  & + \,
1154: \frac{1}{10} \frac{dP_{\rm idd}^{7F}}{dx_F} 
1155: + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1156: isu}^{7F}}{dx_F} + \frac{1}{16} 
1157: \frac{dP_{\rm isu}^{7C}}{dx_F} \right) 
1158: + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1159: isd}^{7F}}{dx_F} + \frac{1}{16} \frac{dP_{\rm isd}^{7C}}{dx_F} \right) 
1160: \nonumber \\ &   & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1161: iss}^{7F}}{dx_F} 
1162: + \frac{3}{17} \frac{dP_{\rm iss}^{7C}}{dx_F} \right) 
1163: %\nonumber \\ &   & + \, 
1164: + \frac{1}{10} \frac{dP_{\rm
1165: iuuu}^{9F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm iuud}^{9F}}{dx_F} 
1166: + \frac{1}{10} \frac{dP_{\rm
1167: iudd}^{9F}}{dx_F} + \frac{1}{10} \frac{dP_{\rm
1168: iddd}^{9F}}{dx_F} \nonumber \\ &   & + \, 
1169: \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1170: isuu}^{9F}}{dx_F} + \frac{1}{28} \frac{dP_{\rm isuu}^{9C}}{dx_F} \right) 
1171: + \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1172: isud}^{9F}}{dx_F} 
1173: + \frac{1}{28} \frac{dP_{\rm isud}^{9C}}{dx_F} \right)
1174: \nonumber \\ &   & + \, 
1175: \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1176: isdd}^{9F}}{dx_F} + \frac{1}{28} \frac{dP_{\rm isdd}^{9C}}{dx_F} \right) +
1177:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1178: issu}^{9F}}{dx_F} + \frac{3}{31} \frac{dP_{\rm issu}^{9C}}{dx_F} \right) 
1179: \nonumber \\ &   & + \, \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1180: issd}^{9F}}{dx_F} + \frac{3}{31} \frac{dP_{\rm issd}^{9C}}{dx_F} \right) +
1181:  \frac{1}{2} \left( \frac{1}{10} \frac{dP_{\rm
1182: isss}^{9F}}{dx_F} + \frac{10}{37} \frac{dP_{\rm isss}^{9C}}{dx_F} 
1183: \right)\,\, , 
1184: \label{xiprobom} 
1185: \\
1186: \frac{dP_{\Omega}^{\rm \Xi^-}}{dx_F} & = & 
1187: \frac{1}{7} \frac{dP_{\rm is}^{5C}}{dx_F} 
1188: + \frac{1}{16} \frac{dP_{\rm isu}^{7C}}{dx_F} 
1189: + \frac{1}{16} \frac{dP_{\rm isd}^{7C}}{dx_F} 
1190: + \frac{3}{17} \frac{dP_{\rm iss}^{7C}}{dx_F}
1191: + \frac{1}{28} \frac{dP_{\rm isuu}^{9C}}{dx_F} 
1192: + \frac{1}{28} \frac{dP_{\rm isud}^{9C}}{dx_F} 
1193: \nonumber \\ &   & + \, 
1194: \frac{1}{28} \frac{dP_{\rm isdd}^{9C}}{dx_F} 
1195: + \frac{3}{31} \frac{dP_{\rm issu}^{9C}}{dx_F} 
1196: + \frac{3}{31} \frac{dP_{\rm issd}^{9C}}{dx_F} 
1197: + \frac{10}{37} \frac{dP_{\rm isss}^{9C}}{dx_F} 
1198: \,\, .
1199: \label{xiprobomco} 
1200: \ee
1201: 
1202: 
1203: \begin{thebibliography}{99}
1204: 
1205: \bibitem{E791}
1206: E.M. Aitala {\it et al.} (E791 Collab.), Phys. Lett. {\bf B371} (1996) 157.
1207: 
1208: \bibitem{ddata1}
1209: M.  Aguilar-Benitez {\it et al.} (LEBC-EHS Collab.),
1210: Phys.  Lett. {\bf 161B} (1985)
1211: 400.
1212: 
1213: \bibitem{ddata2} M.  Aguilar-Benitez {\it et al.} (LEBC-EHS Collab.),
1214: Z.  Phys. {\bf C31} (1986) 491
1215: 
1216: \bibitem{ddata3}
1217: S. Barlag {\it et al.} (ACCMOR Collab.), Z. Phys. {\bf C49} (1991) 555.
1218: 
1219: \bibitem{ddata4}
1220: M.I. Adamovich {\it et al.} (WA82 Collab.), Phys. Lett. {\bf B305} (1993) 402.
1221: 
1222: \bibitem{ddata5}
1223: G.A.  Alves {\it et al.} (E769 Collab.),
1224: Phys.  Rev.  Lett. {\bf 72} (1994) 812.
1225: 
1226: \bibitem{wa891} R. Werding (WA89 Collab.), in Proceedings of {\it ICHEP94},
1227: 27$^{\rm th}$ International Conference on High Energy Physics, Glasgow,
1228: Scotland (1994).
1229: 
1230: \bibitem{wa892}
1231: M.I. Adamovich {\it et al.} (WA89 Collab.), Eur. Phys. J. {\bf C8} (1999) 593.
1232: 
1233: \bibitem{SELEX1} E. Ramberg (SELEX Collab.),
1234: in {\it Hyperons, Charm and Beauty Hadrons},
1235: Proceedings of the 2$^{\rm nd}$ International Conference on Hyperons, Charm and
1236: Beauty Hadrons, Montreal, Canada, 1996, edited by C.S. Kalman {\it et al.},
1237: Nucl. Phys. {\bf B} (Proc. Suppl.) {\bf 55A} (1997) 173. 
1238: 
1239: \bibitem{SELEX2}
1240: J. Engelfried {\it 
1241: et al.}, in Proceedings of the 5$^{\rm th}$ Workshop on Heavy Quarks at Fixed
1242: Target, Rio de Janeiro, Brazil, 2000, hep-ex/0012004.
1243: 
1244: \bibitem{bourq1} M. Bourquin {\it et al.}, Nucl. Phys. {\bf B153} (1979) 13.
1245: 
1246: \bibitem{bourq2} M. Bourquin {\it et al.}, Z. Phys. {\bf C5} (1980) 275.
1247: 
1248: \bibitem{bourq3} M. Bourquin and J.P. Repellin, Phys. Rep. {\bf 114} (1984) 99.
1249:  
1250: \bibitem{biagi} S.F. Biagi {\it et al.}, Z. Phys. {\bf C34} (1987) 187.
1251: 
1252: \bibitem{schn} O. Schneider {\it et al.}, Z. Phys. {\bf C46} (1990) 341.
1253: 
1254: \bibitem{beret} A. Beretvas {\it et al.}, Phys. Rev. {\bf D34} (1986) 53.
1255: 
1256: \bibitem{wa89xi} M.I. Adamovich {\it et al.} (WA89 Collab.), Z. Phys. {\bf 
1257: C76} (1997) 35.
1258: 
1259: \bibitem{Gun} J.F. Gunion, Phys.Lett. {\bf B88} (1979) 150. 
1260: 
1261: \bibitem{Geist} W.M. Geist, Nucl. Phys. {\bf A525} (1991) 149c.
1262: 
1263: \bibitem{Cardello} T.R. Cardello {\it et al.}, Phys. Rev. {\bf D32} (1985) 1.
1264: 
1265: \bibitem{Barton} D.S. Barton {\it et al.}, Phys. Rev. {\bf D27} (1983) 2580.
1266: 
1267: \bibitem{Skubic} P. Skubic {\it et al.}, Phys. Rev. {\bf D18} (1978) 3115.
1268:  
1269: \bibitem{intc1}
1270: S.J.  Brodsky, P.  Hoyer, C.  Peterson and N. Sakai, Phys.  Lett.
1271: {\bf B93} (1980) 451. 
1272: 
1273: \bibitem{intc2} S.J.  Brodsky, C. Peterson and N. Sakai, Phys.
1274: Rev. {\bf D23} (1981) 2745.
1275:  
1276: \bibitem{VB}
1277: R. Vogt and S.J. Brodsky, Nucl. Phys. {\bf B438} (1995) 261.
1278: 
1279: \bibitem{VBlam} R. Vogt and
1280: S.J. Brodsky, Nucl. Phys. {\bf B478} (1996) 311. 
1281: 
1282: \bibitem{GutVogt1} T. Gutierrez and R. Vogt, Nucl. Phys. {\bf B539} (1999) 189.
1283: 
1284: \bibitem{GutVogtIS} T.D. Gutierrez and R. Vogt, Nucl. Phys. {\bf A705} (2002)
1285: 396. 
1286: 
1287: \bibitem{GRV94}  M.~Gl\"uck, E.~Reya, A.~Vogt, Z.~Phys.~{\bf C67}
1288: (1995) 433.
1289:  
1290: \bibitem{GRSpi2}
1291: M. Gl\"{u}ck, E. Reya and I. Schienbein, Eur. Phys. J.
1292: C{\bf 10}, 313 (1999).
1293: 
1294: \bibitem{VBH2}
1295: R.  Vogt, S.J. Brodsky and P. Hoyer, Nucl. Phys. {\bf B383}
1296: (1992) 643.
1297: 
1298: \bibitem{Ellis} R.K. Ellis, in {\it Physics at the 100 GeV Scale}, 
1299: Proceedings of the 
1300: 17$^{\rm th}$ SLAC Summer Institute, Stanford, California, 1989, 
1301: edited by E.C. Brennan (SLAC
1302: Report No. 361) 45.
1303:  
1304: \bibitem{comb} B.L. Combridge, Nucl. Phys. {\bf B151} (1979) 429.
1305: 
1306: \bibitem{NDE} P. Nason, S. Dawson, and R.K. Ellis, Nucl. Phys. {\bf B303}
1307: (1988) 607.
1308: 
1309: \bibitem{hoyerroy} P. Hoyer and D.P. Roy, Phys. Lett. {\bf B410} (1997) 63.
1310: 
1311: \bibitem{VBH1}R. Vogt, S.J. Brodsky, and P. Hoyer, Nucl. Phys. {\bf B360}  
1312: (1991) 67.
1313: 
1314: \bibitem{e789d}M.J. Leitch {\it et al.} (E789 Collab.), Phys. Rev. Lett. 
1315: {\bf 72} (1994) 2542.
1316:  
1317: \bibitem{Badier}
1318: J.  Badier {\it et al.} (NA3 Collab.), Z.  Phys. {\bf C20} (1983) 101.
1319: 
1320: \bibitem{Arneodo} M. Arneodo, Phys. Rep. {\bf 240} (1994) 301.
1321: 
1322: \bibitem{EKS981} K.J. Eskola, V.J. Kolhinen and P.V. Ruuskanen, 
1323: Nucl. Phys. {\bf B535} (1998) 351.
1324: 
1325: \bibitem{EKS982}
1326: K.J. Eskola, V.J. Kolhinen and C.A. Salgado, Eur. Phys. J. {\bf C9} (1999) 61.
1327: 
1328: \bibitem{xiomdat} http://cpt19.dur.ac.uk/cgi-hepdata/hepreac/1637169
1329: 
1330: \end{thebibliography}
1331: 
1332: \end{document}
1333: