hep-ex0303042/cr.tex
1: \documentclass[12pt,a4paper,dvips]{article}
2: \usepackage{a4p}
3: % The default fonts are Computer Modern fonts.
4: % You can get Postscript fonts by uncommenting the line below.
5: % This means the font in the text and figures can be identical.
6: % You have to use dvips -Ppsmath ... if you want to use the psmath package.
7: %\usepackage{times,psmath}
8: \usepackage{cite,mcite}
9: \usepackage{graphicx}
10: % \usepackage{physics }
11: \usepackage{physics}
12: \usepackage{l3_title,ifthen, Lep}
13: %
14: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
15: % Give the journal name
16: \journalname{Phys. Lett. B}
17: % Give the date, when it is final - default is \today
18:  \date{March 13, 2003}
19: % Uncomment the relevant line
20: % \lthreedraft    for drafts
21: % \preprint       for CERN-PPE preprints
22: % \journal        for journal version
23: %                 This also moves the figures to the end on separate
24: %                 pages.
25: %                 Use \journaln to avoid the figures moving macro.
26: %
27: % Draft version:     uncomment the line below and give the version number
28: %\def\lthreenote#1{\def@docversion{{\large L3 Note #1}}}
29: % \lthreenote{2748}
30: % \lthreedraft{4.0}
31: % CERN-PPE preprint: uncomment the line below and give the  preprint number.
32: \preprint{2003-012}
33: % Journal:           uncomment the line below.
34: % \journal
35: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36: %
37: % Indicate the Lep running period by uncommenting one of the lines below.
38: % If you don't, then you will get a warning message when the author list
39: % is included.
40: %
41: %\Lep{1}
42: %\Lep{2}
43: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
44: %
45: % Use \psdraft to avoid printing the figures in early drafts
46: % Use \psfull  to turn printing back on, e.g. if you have one very big
47: % figure that you only want to print on stable versions.
48: %\psdraft
49: %
50: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
51: %
52: % List of directories containing figures.
53: % Each directory must have its own curly brackets {dir1}{dir2}
54: % Don't forget the / at the end of the name.
55: %
56: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
57: %
58: % Use \icaption instead of \caption in tables and figures to get a
59: % caption that is indented by 1cm.
60: % Note that the label should be included inside \icaption for it to
61: % work properly.
62: %
63: \newlength{\capindent}
64: \setlength{\capindent}{1.0cm}
65: \newlength{\capwidth}
66: \setlength{\capwidth}{\textwidth}
67: \addtolength{\capwidth}{-2\capindent}
68: \newlength{\figwidth}
69: \setlength{\figwidth}{\textwidth}
70: \addtolength{\figwidth}{-2.0cm}
71: \newcommand{\icaption}[2][!*!,!]{\hspace*{\capindent}%
72:   \begin{minipage}{\capwidth}
73:     \ifthenelse{\equal{#1}{!*!,!}}%
74:       {\caption{#2}}%
75:       {\caption[#1]{#2}}
76:   \end{minipage}}
77: %
78: 
79: \def\as{\alpha_{\mathrm{s}}}
80: \def\asp{\overline{\alpha}_{\mathrm{s}}}
81: \def\bt{B_{T}}
82: \def\bw{B_{W}}
83: \def\ee{\mathrm{e^{+}e^{-}}}
84: \def\WW{\mathrm{W^{+}W^{-}}}
85: \def\ff{\mathrm{ff}}
86: \def\FF{\mathrm{f} \, {\mathrm{f}}^\prime}
87: \def\etal{{\it et~al.}}%
88: \def\mz{M_{Z}}
89: \def\rs{\sqrt{s}}
90: \def\rsp{\sqrt{s'}}
91: \def\yc{y_{cut}}
92: \def\qq{\mathrm{q}\bar{\mathrm{q}}}
93: \def\ksistar{\xi^\star}
94: \def\rspn{\sqrt{s'/s}}
95: \def\Nf{\mathrm{N_{f}}}
96: \def\Nc{\mathrm{N_{c}}}
97: \def\nch{\langle \mathrm{n_{ch}} \rangle}
98: \def\WW{\mathrm{W^{+}W^{-}}}
99: \def\TT{\tau^{+}\tau^{-}}
100: \def\WW{\mathrm{W^{+}W^{-}}}
101: \def\Z{\mathrm{Z}}
102: \def\ZZ{\mathrm{Z} \mathrm{Z}}
103: 
104: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
105: % This is where the document really begins
106: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
107: %
108: \begin{document}
109: \bibliographystyle{l3style}
110: %       
111: \begin{titlepage}
112: %
113: 
114: \title{ Search for Colour Reconnection Effects in \\
115:  \boldmath$\ee \rightarrow \WW \rightarrow \mathrm{hadrons}$
116:  through  Particle-Flow Studies at LEP}
117: 
118: 
119: \author{The L3 Collaboration}
120: 
121: %
122: % The abstract
123: %
124: \begin{abstract}
125: A search for colour reconnection effects in 
126:  hadronic decays of W pairs is performed with the L3 detector at centre-of-mass energies
127:  between 189 and 209 \GeV. 
128:  The analysis is based on 
129:   the study of the particle flow  between jets associated to the same W boson and between two 
130:  different W bosons in $\qq\qq$ events.
131:  The ratio of particle yields in the different  interjet regions
132: is found to be sensitive to colour reconnection effects implemented in some 
133: hadronisation models. The data are compared to different models with and
134: without such effects. An extreme scenario  of colour reconnection is ruled out. 
135: 
136: \end{abstract}
137: %
138: % Adds "To be submitted to ..." or "Submitted to ...", if relevant
139: %
140: \submitted
141: % \vspace*{20mm}
142: % \centerline{To be submitted to Phys. Lett. B. }
143: 
144: 
145: 
146: \end{titlepage}
147: 
148: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
149: \vspace{5cm}
150: 
151: \section{Introduction}
152: 
153: According to the string model of hadronisation, the particles produced 
154: in the process  $\ee\rightarrow\WW\rightarrow hadrons$
155: originate, in  the absence of colour reconnection,
156:  from the fragmentation of two colour singlet strings each of
157:  which is
158: stretched between the two quarks from a W boson. 
159: In this case the hadrons are uniquely associated to a particular W and  
160:  there is a direct correspondence between the  jets formed by these
161: hadrons and the primary quarks from the W boson decays. 
162:  Energy-momentum is separately conserved for each of the W systems.
163: However,
164: it has been suggested that interactions may occur between the decay products of
165: the two W bosons~\cite{gusta,sjost,gh,yellow}.
166: The main justification for this ``cross-talk'' is the relatively short distance separating the
167: decay vertices of the W bosons produced  in $\ee$ annihilation
168: ($\approx$ 0.1 fm) compared to the typical hadronic scale (1 fm), 
169: which implies a large space-time overlap of the two hadronising systems.
170: 
171: The main consequence of these interactions, called Colour Reconnection (CR)
172: effects, is a modification of the distribution in phase space of hadrons.
173: CR effects are thought to be suppressed in the
174: hard perturbative phase, but may be more important in the soft gluon 
175: emission regime~\cite{sjost}.
176:  While hard gluons, with energy greater than the W width, are radiated independently
177: from different colour singlets,
178:   soft gluons could in principle  be affected by the colour strings of both decaying W's.
179: Such CR would affect the number of  soft particles in specific
180:  phase space regions, especially  outside the jet cores.
181: 
182: The study of CR is interesting not only for probing  QCD dynamics but also 
183: for determining a possible bias in the W mass measurement
184: in the four-quark channel.
185: CR could affect the invariant masses of jet pairs originating from
186: W decays. Therefore
187:  the precision with which the W mass may be determined using the four-quark channel
188: depends strongly on the understanding of CR effects.
189: Events where only one W decays hadronically are unaffected by CR.
190: 
191: Previous LEP  studies of CR, performed at centre-of-mass energy $\rs \le$ 183 \GeV,  
192: were based on charged particle multiplicity and momentum 
193: distributions~\cite{lepnchcr}.
194: 
195: The analysis presented in this paper uses the 
196: method suggested in Reference~\citen{crdom} based on energy and particle flow
197: to probe the string topology of four-quark events to search for  particular
198:  effects of particle depletion and  enhancement.
199: The results are based on 627 pb$^{-1}$ of data collected 
200: with the L3 detector~\cite{l3detect} at  $\rs$=189$-$209 \GeV.
201: Comparisons with various  models are made at detector level and the compatibility with
202: the existence of CR effects in various models is investigated.
203:  
204: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
205: \section{Colour Reconnection Models}
206: Several phenomenological models have been 
207: proposed~\cite{sjost,gh,geiger,arcr,hwcr,rathsman}
208: to describe  CR effects in  $\ee\rightarrow\WW\rightarrow hadrons$
209:  events.
210: The analysis presented in this paper is performed with  some of those
211: CR models, which are
212: implemented in the PYTHIA~\cite{pythia}, ARIADNE~\cite{ariadne}
213: and HERWIG~\cite{herwig} Monte Carlo (MC)  programs. 
214: 
215: We investigate two models by  Sj\"ostrand and Khoze~\cite{sjost} implemented
216: in PYTHIA. They are based on
217: rearrangement of the string configuration during the fragmentation
218: process. They follow
219: the space-time evolution of the strings and allow local reconnections
220: if the strings overlap or cross,
221:  depending on the string definition (elongated bags  or vortex lines).
222: 
223: In the type I model (SKI)  the strings are associated with colour flux tubes having a
224:  significant transverse extension. The  reconnection occurs when these tubes 
225:  overlap and only one reconnection
226:   is allowed, the one with the largest overlap volume. The reconnection
227:    probability depends on this  volume of overlap and is controlled by 
228:    one free parameter, $k_{\rm{I}}$, which can be varied in the model 
229: to generate event samples  with different fractions of reconnected events.
230: The relation with the event reconnection probability ($P_{\rm{reco}}$) is given by 
231: the following formula:
232: \begin{eqnarray}
233: P_{\rm{reco}} = 1-{\rm{exp}}(-f k_{\rm{I}})
234: \end{eqnarray}
235: where $f$ is a function of the overlap volume of the two strings, which 
236: depends on W-pair kinematics varying with $\rs$.
237: The default value of $k_{\rm{I}}$ is 0.6~\cite{sjost}, 
238: which corresponds
239: to  a reconnection probability of about 30\% at $\rs$=189 \GeV. 
240: This analysis is performed with three different values of  
241:  $k_{\rm{I}}$: 0.6, 3 and 1000, corresponding to reconnection probabilities
242:  at  $\rs$=189 \GeV\ of about 30\%, 66\% and nearly 100\%, respectively.
243: 
244:  In the type II model (SKII)  the strings have no lateral extent
245: and  the  reconnection occurs, with unit probability, when they cross. The 
246: fraction of reconnected events in this model is of the order of 30\%
247:  at $\rs$=189 \GeV.
248: 
249:  The  CR model  implemented in ARIADNE is based on reconnection
250: of  coloured dipoles before the string fragmentation takes place~\cite{arcr}.
251:    In the AR2 scheme, which is investigated here, reconnections are allowed
252: if they reduce the string length.  While reconnections within a W are
253: allowed at all scales, those between W's are only allowed after the parton
254: showers have evolved down to gluon energies less than 2 \GeV.
255:  At $\rs=$189 \GeV\ they affect about 55\% of the events.
256: 
257: The CR scheme implemented in HERWIG is, as for the string fragmentation, a local
258: phenomenon since the cluster fragmentation process follows the space-time 
259: development. In this model~\cite{hwcr} the clusters are rearranged if their space-time
260: extension is reduced. This rearrangement occurs with a probability equal
261: to 1/$N^{2}_{\rm{colour}}$, with default value $N_{\rm{colour}}$ = 3, giving
262: about 23\% of reconnected events.
263: 
264: All probabilities discussed above are derived as fraction of events where at
265: least one reconnection occurs either within the same W or between two W's.
266: 
267: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
268: 
269: 
270: \section{Event Selection}
271: 
272: The energy measured 
273: in the electromagnetic and hadronic calorimeters and in the tracking chamber
274: is used to select  $\ee\rightarrow\WW\rightarrow hadrons$ events. 
275:  The total visible energy 
276: (${E_{\mathrm{vis}}}$) and the energy imbalance parallel (${E_{\parallel}}$)
277: and perpendicular (${E_{\perp}}$) to the beam direction are measured. 
278:  The number of clusters, defined as objects obtained from a non-linear 
279: combination of charged tracks with a transverse momentum greater than 100 \MeV\
280: and calorimetric clusters 
281: with a minimum energy of 100 \MeV, is denoted  by 
282: ${N_{\mathrm{cluster}}}$.
283: The selection criteria are:
284: \vspace*{-0.1cm}
285: \begin{center}
286:    ${E_{\mathrm{vis}}}/ \rs > 0.7$; 
287:     \hspace*{1cm}  ${E_{\perp} / E_{\mathrm{vis}}}  < 0.2$; 
288:   \hspace*{1cm}  ${|E_{\parallel}| / E_{\mathrm{vis}}}  < 0.2$;
289:    \hspace*{1cm}  ${N_{\mathrm{cluster}}} \geq 40 $.
290: \end{center}   
291: In addition the events must have 4 jets reconstructed with the Durham algorithm~\cite{kt} 
292: with $y_{\rm{cut}}$ = 0.01.
293: To reduce the contamination from
294: semileptonic W decays, events with energetic  $\mu$ or $\rm{e}$  are rejected.
295: Events with hard initial state radiation (ISR) are rejected as described in
296: Reference~\citen{qcdlep2}.
297: Additional criteria select events with nearly perfect quark-jet association,
298: necessary for the study of particle and energy flow between jets. 
299: The two largest interjet angles are required to be between $100^{\circ}$ and $140^{\circ}$ and
300:  not adjacent.
301: The two other interjet angles must be less than $100^{\circ}$.
302: This selection guarantees 
303: similar sharing of energy between the four primary partons with the two strings 
304: evolving back-to-back and similar interjet regions between the two W's.
305: The above cuts are optimized by studying MC $\WW$ events at $\rs$=189
306: \GeV\ using the KORALW~\cite{koralw} MC generator interfaced with the PYTHIA 
307: fragmentation model without CR.
308: Relaxing the angular criteria increases the efficiency 
309: but gives lower probability to have correct 
310:  W-jet pairing  due to the more complicated event topology.
311: 
312: The number of selected events, the number of expected events, the selection
313: efficiency and the percentage of correct pairing are given in
314: Table~\ref{tab:selflow}.
315: After applying all the cuts the full sample contains 
316: 666 events with an average efficiency of 12\% and a purity of about 85\%
317: for $\ee\rightarrow\WW\rightarrow hadrons$.
318: The average probability to have the correct pairing between the W bosons and their
319:  associated  jets  is estimated to be 91\%.
320: 
321: The  background is composed of  $\qq (\gamma)$ events
322: and   Z-pair production events, in similar amounts. Background from semileptonic W pair decays
323: is found to be negligible (less than 0.3\%).  The $\qq (\gamma)$ 
324:  process is  modeled with the KK2F MC program~\cite{kk2f}, interfaced
325: with JETSET~\cite{jetset} routines to describe the QCD processes, and
326: the background from Z-pair production is simulated with PYTHIA.
327:   For CR studies W-pair events are simulated with PYTHIA.
328: All MC samples  are passed through a realistic detector simulation~\cite{l3-simul}
329: which takes into account  time dependent detector effects and  inefficiencies. 
330: 
331: 
332: 
333: \section{Particle- and Energy-Flow Distributions}
334: The algorithm  to build the particle- and energy-flow distributions~\cite{crdom} 
335: (Figure~\ref{fig:plane}) starts by
336:  defining the plane spanned by the most energetic jet (jet 1) and
337: the  closest jet making an angle with jet 1 greater than  $100^{\circ}$
338:  which is most likely associated to the same  W (jet 2).
339: For each event, the momentum vector  direction of each  particle is then
340: projected on to this plane.
341: The particle and energy flows are measured as a function of the angle, $\phi$, 
342:  between 
343: jet 1 and the projected momentum vector for the particles located between jets 1 and 2.  
344: In order to take into account the fact that the W-pair events are not   planar 
345: a new plane is defined for each remaining pair of adjacent jets.
346: In this  four-plane configuration the angle $\phi$ is defined as increasing from jet 1 toward jet
347: 2, then to the closest jet from the other W (jet 3) toward the 
348: remaining jet (jet 4) and back to jet 1. 
349: The angle $\phi_{j,i}$ of a  particle $i$ having a projected momentum vector located
350:  between  jets $j$ and $j+1$
351: is calculated in the plane spanned by these two jets. 
352: A particle $i$ making an angle $\phi_{i}$ with respect to jet 1 adds an entry equal to 1
353: in the particle-flow distribution and adds an entry equal to its energy, normalised
354:  to the total event energy,
355: in the energy-flow distribution for  the corresponding $\phi$ bin.
356:   
357: The distributions are calculated using, for the particle definition,
358:  the clusters defined in the previous section.
359: 
360:  Figure~\ref{fig:flow2} shows the particle- and energy-flow distributions obtained
361:  for the data and the MC
362: predictions at detector level by using only the first plane for projecting all the
363: particles.
364: The data and MC distributions agree over the full angular range in both cases. 
365: 
366: In order to compare the interjet regions 
367: the angles in the planes are rescaled by the angle between the two closest jets.
368: For a particle $i$ located between jets $j$ and $j+1$ the rescaled 
369: angle is
370: \begin{eqnarray}
371:  \phi_{i}^{\mathrm{resc}} = j - 1 +\frac{\phi_{j,i}}{\psi_{j,j+1}}
372: \end{eqnarray}
373:  where $\phi_{j,i}$ is the angle between jet $j$ and particle $i$ and  $\psi_{j,j+1}$
374:  is the angle between jets $j$ and  $j+1$.
375:  With this definition the four jets have fixed rescaled angle values
376:  equal to 0, 1, 2 and 3.
377:  
378: Figure~\ref{fig:flow3}a shows the rescaled particle-flow distribution normalised to the
379: number of events
380:    after a bin-by-bin  background
381: subtraction for the data  and  MC predictions 
382: without CR  and for the SKI model with $k_{\rm{I}}$=1000, later referred to as SKI 100\%.
383: As expected, the latter shows some depletion in the number of particles  in 
384: the intra-W regions spanned by the two W bosons (regions A and B)
385:  and some particle enhancement in the two inter-W regions (regions C and D)
386:  when compared to the model without CR (no-CR).
387: 
388: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
389: To improve the sensitivity to CR effects the particle flows in  regions A and B 
390:   are averaged as are the particle flows in regions C and D.
391: The results are shown in  
392: Figures~\ref{fig:flow3}b and \ref{fig:flow3}c where the angle is redefined to be in the range [0,1].
393: MC studies at particle level 
394:  with particles  having a momentum greater than 100 \MeV\
395: show that the CR effects are consistent with the detector level results
396: and have similar magnitudes.
397: 
398: The ratio of the particle flow between the quarks from the 
399: same W  to that between quarks from  different W's  
400: is found to be a sensitive observable to 
401: cross-talk effects as predicted by the SKI model.
402: These ratios, computed from the particle- and energy-flow distributions at 
403: detector level, are shown in Figure~\ref{fig:flowrat} for the data,
404:  the PYTHIA prediction without CR, the SKI model with 
405: $k_{\rm{I}}$=3 and SKI 100\%.
406: 
407: The differences between the models with and without CR
408: are larger in the middle of the interjet regions. 
409: Therefore, in order to quantify the CR effects the 
410: ratio $R$ is computed
411: in an  interval, 0.2  $< \phi_{\mathrm{resc}} <$ 0.8, optimized with respect to the
412: sensitivity to SKI 100\%. 
413: The  corresponding variables for
414: particle and energy flow  are defined as follows:
415: 
416: \begin{eqnarray}
417: R_{\rm{N}}=  \int_{0.2}^{0.8}{f_{\rm{N}}^{\rm{A+B}}}
418: \hspace*{0.3cm}{\rm{d}}\phi \left/\ \int_{0.2}^{0.8}{f_{\rm{N}}^{\rm{C+D}}}
419: \hspace*{0.3cm}{\rm{d}}\phi 
420: \hspace*{0.3cm}\mathrm{ and}\hspace*{0.3cm}
421: R_{\rm{E}}=  \int_{0.2}^{0.8}{f_{\rm{E}}^{\rm{A+B}}}
422: \hspace*{0.3cm}{\rm{d}}\phi \right/\ \int_{0.2}^{0.8}{f_{\rm{E}}^{\rm{C+D}}}
423: \hspace*{0.3cm}{\rm{d}}\phi 
424: \end{eqnarray}
425: where, in a region $i$,
426: \begin{eqnarray}
427: f_{\rm{N}}^{i}= \frac{1}{N_{\rm{evt}}}\frac{{\rm{d}}n}{{\rm{d}}\phi}
428: \hspace*{0.3cm}\mathrm{ and}\hspace*{0.3cm}
429: f_{\rm{E}}^{i} = \frac{1}{E}\frac{{\rm{d}}E}{{\rm{d}}\phi}
430: \end{eqnarray}
431: 
432: The measured values of $R_{\mathrm{N}}$ and $R_{\mathrm{E}}$ obtained at each centre-of-mass 
433: energy are summarised in
434: Table~\ref{tab:result}. Correlations
435: in the particle rates between the four interjet regions are
436: taken into account by constructing the full covariance matrix.
437:  This results in an increase of about 20\% of the statistical uncertainty.
438: The  values obtained with the complete data sample are:
439: \begin{eqnarray*}
440: R_{\rm{N}}  =  0.911 \pm 0.023\hspace*{0.1cm}(\rm{stat.})\\
441: R_{\rm{E}}  =  0.719 \pm 0.035\hspace*{0.1cm}(\rm{stat.}) 
442: \end{eqnarray*}
443: An estimate of the sensitivity to the SKI 100\% model, shows that  $R_{\rm{N}}$
444:  is 2.6 times more sensitive than $R_{\rm{E}}$. 
445: Accordingly, the following results and discussion are only based on $R_{\rm{N}}$.
446: 
447: Figure~\ref{fig:ratio} shows the measured $R_{\rm{N}}$ as a function of
448: $\rs$ together  with PYTHIA no-CR and SKI model predictions.
449:  The energy dependence originating from  the 
450: different pairing  purities 
451: and jet configurations is in agreement with the model predictions.
452: For the PYTHIA SKI predictions,
453: the ratio decreases with the reconnection probability over the whole 
454: energy range with similar magnitude. 
455: The data indicate little or no CR.
456: 
457: \section{Semileptonic Decays}
458: To verify the quality of the MC simulation of the $\mathrm{W}\rightarrow\qq$ 
459: fragmentation process and the possible biases
460:  which may arise when determining the particle yields between reconstructed jets in the detector, 
461:  the particle- and energy-flow distributions are investigated in
462:  $\ee\rightarrow\WW\rightarrow\qq l\nu$ where $l=\mathrm{e},\mu$.
463: For this analysis events are selected with high multiplicity,
464: large missing momentum and a high energy electron or muon.
465: The missing momentum is considered as a fictitious particle in order to apply the Durham
466:  jet algorithm to select  4-jet events with  $y_{\rm{cut}}$=0.01.
467: 
468:   The same angular criteria  on the four interjet angles 
469:  as applied in the fully hadronic channel are used here. 
470: The purity obtained after selection is about 96\% and the efficiency 
471:  is about 12\%. The number of selected semileptonic
472: events  is 315 with an expectation of 314.5 events.
473: Particle- and energy-flow distributions are built in a similar way as
474:  in the fully hadronic channel with the additional requirement that 
475:  the charged lepton should be in  jet 3 or 4. 
476: Figure~\ref{fig:flowlept}a
477: shows the corresponding particle-flow distribution projected on to the plane of 
478: jets 1 and 2 for  the data 
479: and the KORALW MC prediction. There is good agreement between data and MC
480: over the whole distribution. 
481: Figure~\ref{fig:flowlept}b shows the rescaled particle-flow distribution where 
482: the  structure of 
483: the two different W's is clearly visible.
484: The region between jet 1 and jet 2 corresponds to the hadronically decaying W ($W_{1}$) and
485: the region between jet 3 and jet 4 corresponds to the W decaying semileptonically ($W_{2}$).
486:  The  activity in the $W_{2}$ region
487: is mainly due to low energy fragments from the hadronic decay of the first W.
488: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
489: A comparison of data and MC for the particle flow obtained by summing the regions 
490:  $W_{1}$ and $W_{2}$  is shown in Figure~\ref{fig:flowlept2}a.
491: The ratio between the data and the MC distributions is  shown in Figure~\ref{fig:flowlept2}b.
492: This ratio is consistent with unity over the whole range.
493: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
494: This result gives additional confidence in the correctness of the
495:  modelling of the fragmentation process of  quark pairs according to the 
496:  fragmentation parameters used in  KORALW and PYTHIA 
497:  as well as the particle flow definition and reconstruction.
498: 
499: In the absence of CR effects, the  activity found in regions A+B of  a 
500: fully hadronic event should be equivalent to  twice the particle activity in the 
501: regions  $W_{1}$+$W_{2}$  of the distribution for a semileptonic event. Figure~\ref{fig:flowlept2}c
502: shows the ratio of the particle flow in four-quark events divided by twice the particle flow 
503: in semileptonic events for the data  and the predictions from no-CR PYTHIA MC  and
504:  the SKI 100\% model. The CR model shows the expected deficit in the hadronic
505:  channel compared to the semileptonic one. The data are consistent with the no-CR scheme but the 
506: large statistical uncertainty prevents a quantitative
507:  statement based on this model-independent comparison. 
508: 
509: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
510: \section{Systematic Uncertainties}
511: 
512: Several sources of systematic uncertainties are investigated. 
513: The first  important  test is whether the
514: result depends on  the definition of the particles. The analysis is 
515: repeated using calorimetric clusters only.
516:  Half  the difference between the two analyses is assigned as the uncertainty due
517: to this effect. This is found to be the dominant 
518: systematic uncertainty.
519: 
520: The second source of systematic uncertainty 
521: is the limited knowledge of  quark fragmentation modelling.
522: The systematic effect in the $\qq (\gamma)$ background 
523:  is estimated by comparing results using 
524: the JETSET and HERWIG MC programs.  The corresponding uncertainty is
525: assigned to be half  the  difference between the two models.
526: 
527: The systematic uncertainty from quark fragmentation modelling in W-pair events
528: is estimated by comparing results using
529: PYTHIA, HERWIG and ARIADNE MC samples without CR. The uncertainty is assigned as
530: the RMS  between the $R_{\mathrm{N}}$ values obtained with the three fragmentation models.
531: Such comparisons between different models  test also possible effects
532: of different fragmentation schemes
533: which are not taken into account when varying only fragmentation parameters
534: within one particular model.
535:  
536: Another source associated with fragmentation modelling  is
537: the  effect of   Bose-Einstein correlations (BEC) in hadronic
538: W decays. This effect  is estimated by repeating the analysis using
539: a MC sample with BEC only between particles originating
540: from the same W. An uncertainty is assigned equal to half the
541: difference with the default MC which includes full BEC simulation (BE32 option)~\cite{be32} 
542: in W pairs. 
543: The sensitivity of the $R_{\rm{N}}$ variable to BEC is found to be small.
544: 
545: The third main source of systematic uncertainty is the background estimation.
546: The  $\qq (\gamma)$ background which is subtracted corresponds mainly to QCD four-jet 
547: events for which the rate is not well modelled by parton shower programs. PYTHIA underestimates,
548: by about 10\%, the four-jet rate in the selected phase space region~\cite{4jetlep}.
549: A systematic uncertainty is
550:  estimated by varying the  $\qq (\gamma)$  cross section by $\pm$ 5\% after correcting
551:  the corresponding background by +5\%.
552:   This correction increases the value of $R_{\rm{N}}$
553:   by 0.004.
554: 
555: A last and small systematic uncertainty is associated with  Z-pair production. 
556: It is estimated by varying the corresponding cross section by $\pm$ 10\%.
557: This variation takes into account all possible uncertainties pertaining to
558: the hadronic channel, from final state interaction effects to the theoretical 
559: knowledge of the hadronic cross section.
560: 
561: A summary of the different contributions to the systematic uncertainty is given
562: in Table~\ref{tab:syst}.
563: 
564: The ratio   obtained by taking into account the systematic uncertainties  is then:
565: \begin{eqnarray*}
566: R_{\rm{N}}  =  0.915 \pm 0.023 \hspace*{0.1cm} (\rm{stat.}) \pm 0.021 \hspace*{0.1cm}(\rm{syst.})
567: \end{eqnarray*}
568: 
569: 
570: \section{Comparison with Models}
571: 
572:  The $R_{\rm{N}}$ values predicted by the PYTHIA no-CR, SKI,  SKII, ARIADNE no-CR, 
573: AR2, HERWIG no-CR and HERWIG CR  models are given in Table~\ref{tab:model}.
574: 
575:   The data disfavour extreme scenarios of CR.
576:  A comparison with ARIADNE and HERWIG  shows that
577:  the CR schemes implemented in these two models 
578:  do not modify significantly the interjet particle activity in the hadronic W-pair decay events.
579:   Thus it is not possible to constrain either of these models in the present analysis.
580: 
581: The dependence of $R_{\rm{N}}$ 
582:  on the reconnection probability is investigated with the SKI model.
583: For this, four MC samples are used: the no-CR sample and those 
584:  with $k_{\rm{I}}$=0.6, 3 and 1000. 
585: In the SKI model the fraction of reconnected events is controlled by the $k_{\rm{I}}$
586: parameter and the dependence of $R_{\rm{N}}$ on $k_{\rm{I}}$ is parametrized as
587: $R_{\rm{N}}(k_{\rm{I}}) = p_{1} (1-{\rm{exp}}(-p_{2} k_{\rm{I}}))+p_{3}$ where
588:  $p_{i}$ are   free parameters.
589:  A $\chi^{2}$ fit to the data is performed.
590: The $\chi^{2}$ minimum is at $k_{\rm{I}}=0.08$.  This value corresponds to about 6\%
591: reconnection probability  at $\rs$=189 \GeV. Within the
592:  large uncertainty  the result is also consistent with no CR effect. 
593: 
594: The upper limits  on $k_{\rm{I}}$ at 68\% and 95\% confidence level are derived
595: as 1.1 and 2.1 respectively.
596: The corresponding reconnection probabilities at $\rs$ = 189 \GeV\ are 45\% and 64\%.
597: The extreme SKI scenario, in which CR
598: occurs in essentially all events, is disfavoured by 4.9 $\sigma$.
599: 
600: 
601: 
602: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
603: %
604: %\clearpage
605: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
606: % Bibliography
607: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
608: %
609: % Style file to use with mcite.
610: % Use l3style with just cite.
611: % \bibliographystyle{/l3/paper/biblio/l3stylem}
612: % \bibliography{l3at183}
613: \begin{thebibliography}{99}
614: 
615: \bibitem{gusta} G. Gustafson, U. Pettersson and P. Zerwas,
616: Phys. Lett. {\bf B 209} (1988) 90.
617: 
618: \bibitem{sjost} T. Sj\"ostrand and V.A. Khoze, Phys. Rev. Lett. {\bf 72} (1994) 28;
619:  Z. Phys. {\bf C 62} (1994) 281;
620: V.A. Khoze and T. Sj\"ostrand, Eur. Phys. J. {\bf C 6} (1999) 271.
621: 
622: \bibitem{gh} G. Gustafson and J. H\"akkinen, Z. Phys. {\bf C 64} (1994) 659.
623: 
624: \bibitem{yellow} A. Ballestrero et al.\ in ``Physics at LEP2'', 
625: eds. G. Altarelli et al., CERN 96-01 (1996) 141.
626: 
627: \bibitem{lepnchcr}
628: OPAL Collab., K. Ackerstaff \etal, Eur. Phys. J. {\bf C 1} (1998) 395;
629:  G. Abbiendi \etal, Phys. Lett.  {\bf B 453} (1999) 153;\\
630: DELPHI Collab., P. Abreu \etal, Eur. Phys. J. {\bf C 18} (2000) 203;
631: Erratum {\em ibid.} {\bf C 25} (2002) 493.
632: 
633: \bibitem{crdom}
634: D. Duchesneau,  preprint LAPP-EXP 2000-02, (2000)\\
635: available at http://wwwlapp.in2p3.fr/preplapp/psexp/lappexp0002.ps.gz
636: 
637: \bibitem{l3detect} L3 Collab, B. Adeva \etal, Nucl. Instr. and Meth.
638: {\bf A 289} (1990) 35; \\
639: M. Chemarin  \etal, Nucl. Instr. and Meth. {\bf A 349} (1994) 345; \\
640: M. Acciarri  \etal, Nucl. Instr. and Meth. {\bf A 351} (1994) 300; \\
641: G. Basti  \etal, Nucl. Instr. and Meth. {\bf A 374} (1996) 293; \\
642: I. C. Brock \etal, Nucl. Instr. and Meth. {\bf A 381} (1996) 236; \\
643: A. Adam  \etal, Nucl. Instr. and Meth. {\bf A 383} (1996) 342.
644: 
645: \bibitem{geiger}
646: J. Ellis and K. Geiger, Phys. Rev. {\bf D 54} (1996) 1967.
647: 
648: \bibitem{arcr}
649: L. L\"onnblad,  Z. Phys. {\bf C 70} (1996) 107.
650: 
651: \bibitem{hwcr}
652: B. Webber, J. Phys. {\bf G 24} (1998) 287.
653: 
654: \bibitem{rathsman}
655: J. Rathsman, Phys. Lett. {\bf B 452} (1999) 364.
656: 
657: \bibitem{pythia} 
658:      PYTHIA 6.1 Monte Carlo Program:\\
659:       T. Sj\"ostrand \etal, Comp. Phys. Comm. {\bf 135} (2001) 238.
660: 
661:  \bibitem{ariadne}
662:      ARIADNE 4.08  Monte Carlo Program: \\
663:      L. L\"onnblad, Comp. Phys. Comm. {\bf 71} (1992) 15.
664: 
665:  \bibitem{herwig}
666:      HERWIG 6.2 Monte Carlo Program: \\
667:      G. Corcella  \etal, JHEP {\bf 01} (2001) 010;\\
668:     G. Marchesini \etal, Comp. Phys. Commun. {\bf 67} (1992) 465. 
669: 
670:   \bibitem{kt}
671:      S. Catani \etal, Phys. Lett. {\bf B 269} (1991) 432; \\
672:      N. Brown and W.J. Stirling, Z. Phys. {\bf C 53} (1992) 629; \\
673:      S. Bethke \etal, Nucl. Phys. {\bf B 370} (1992) 310; erratum: ibid. {\bf B 523} (1998) 681.
674:  
675:   \bibitem{qcdlep2}
676:  L3 Collab., P. Achard \etal, Phys. Lett.  {\bf B 536} (2002) 217.
677:  
678:  \bibitem{koralw} 
679:      KORALW 1.42 Monte Carlo Program:\\
680:      S. Jadach \etal, Comp. Phys. Comm. {\bf 119} (1999) 272.
681:  
682: \bibitem{kk2f} 
683:      KK2F 4.14 Monte Carlo Program:\\
684:      S. Jadach, B. F. L. Ward and Z. W\c{a}s, Phys. Lett. {\bf B 449} (1999) 97.
685: 
686:  \bibitem{jetset}
687:       JETSET 7.4  Monte Carlo Program: \\
688:      T. Sj\"ostrand, preprint CERN-TH-7112/93 (1993), revised 1995;\\
689:      T. Sj\"ostrand, Comp. Phys. Comm. {\bf 82} (1994) 74.
690:  
691:   \bibitem{l3-simul} 
692:      The L3 detector simulation is based on GEANT Version 3.15.\\
693:      R. Brun \etal, preprint CERN-DD/EE/84-1 (1984), revised 1987.\\
694:      The GHEISHA program (H. Fesefeldt, RWTH Aachen Report PITHA 85/02 
695:      (1985)) is used to simulate hadronic interactions.
696:  
697: \bibitem{be32}
698: L. L\"onnblad and  T. Sj\"ostrand, Eur. Phys. J. {\bf C 2} (1998) 165.
699: 
700: \bibitem{4jetlep}
701: A. Ballestrero \etal, preprint hep-ph/0006259 (2000).
702:  
703: \end{thebibliography}
704: 
705: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
706: % The author list
707: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
708: %
709: \newpage
710: %\section*{Author List}
711: \input namelist266.tex
712: \newpage
713: %%
714: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
715: %  TABLES
716: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
717: \begin{table}[htb]
718: \begin{center}
719: \begin{tabular}{|c|c|c|c|c|c|}
720: \hline
721: $\langle\rs\rangle$ (GeV)  &  $ {\cal{L}} (\rm{pb}^{-1})$ & $N_{\rm{events}}$ & $N_{\rm{MC}}$ & $\epsilon$ & $\pi$  \\
722: \hline
723:  \hline
724: 188.6 & 176.7 & 208 & 226.0 & 14.2\% & 88\%\\
725: \hline
726: 191.6 & \phantom{1}29.7 &  \phantom{1}38 &  \phantom{1}37.9 & 14.3\% & 90\%\\
727: \hline
728: 195.5 & \phantom{1}83.7 & 104 & 101.0 & 13.4\% & 92\%\\
729: \hline
730: 199.5 &  \phantom{1}84.3 &  \phantom{1}97 &  \phantom{1}91.9 & 12.2\% & 93\%\\
731: \hline
732: 201.7 &  \phantom{1}35.5 &  \phantom{1}36 &  \phantom{1}37.2 & 11.3\% & 93\%\\
733: \hline
734: 205.1 &  \phantom{1}77.8 &  \phantom{1}75 &  \phantom{1}74.8 & 10.3\% & 93\%\\
735: \hline
736: 206.6 & 138.9 & 108 & 120.8 &  \phantom{0}8.9\% & 91\%\\
737: \hline
738: \hline
739: 198.2 & 626.6 & 666 & 689.6 & 12.0\% & 91\%\\
740: \hline
741: \end{tabular}
742: \caption[]{Average centre-of-mass energies, integrated luminosities (${\cal{L}}$), number of selected
743: events ($N_{\rm{events}}$), number of expected events ($N_{\rm{MC}}$), selection efficiency ($\epsilon$)
744: and percentage of
745: correct jet pairing ($\pi$) for the particle flow analysis. The combined figures are given in
746: the last row.}
747: \label{tab:selflow}
748: \end{center}
749: \end{table}
750: 
751: \begin{table}[htb]
752: \begin{center}
753: \begin{tabular}{|c|c|c|}
754: \hline
755: $\langle\rs\rangle$ (GeV)  &  $ R_{\rm{N}}$ & $R_{\rm{E}}$  \\
756: \hline
757:  \hline
758: 188.6 & 0.820 $\pm$ 0.037 & 0.610 $\pm$ 0.047\\
759: \hline
760: 191.6 & 0.929 $\pm$ 0.093 & 0.822 $\pm$ 0.133 \\
761: \hline
762: 195.5 & 0.948 $\pm$ 0.059 & 0.774 $\pm$ 0.077 \\
763: \hline
764: 199.5 & 1.004 $\pm$ 0.067 & 0.871 $\pm$ 0.095 \\
765: \hline
766: 201.7 & 0.770 $\pm$ 0.086 & 0.626 $\pm$ 0.130 \\
767: \hline
768: 205.1 & 1.033 $\pm$ 0.083 & 0.756 $\pm$ 0.111 \\
769: \hline
770: 206.6 & 0.958 $\pm$ 0.068 & 0.781 $\pm$ 0.096 \\
771: \hline
772: \end{tabular}
773: \caption[]{Measured $R_{\rm{N}}$ and $R_{\rm{E}}$ values as a function of energy with
774: their statistical uncertainties.}
775: \label{tab:result}
776: \end{center}
777: \end{table}
778: 
779: \begin{table}[htb]
780: \begin{center}
781: \begin{tabular}{|c|c|}
782: \hline
783: Source    & $\sigma_{R_{\rm{N}}}$  \\
784: \hline
785:  \hline
786: Energy flow objects  &    0.016\\
787: \hline
788: $\qq$  fragmentation  &     0.009\\
789: \hline
790: WW fragmentation   &   0.008 \\
791: \hline
792: BEC  &     0.003\\
793: \hline
794:  4-jet background rate &    0.004\\
795: \hline
796: ZZ background  &     0.002\\
797: \hline
798: \hline
799: Total  &  0.021\\
800: \hline
801: \end{tabular}
802: \caption[]{Contributions to the systematic uncertainties on  $R_{\rm{N}}$.}
803: \label{tab:syst}
804: \end{center}
805: \end{table}
806: 
807: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
808: \begin{table}[htb]
809: \begin{center}
810: \begin{tabular}{|c|c|}
811: \hline
812:    & $R_{\rm{N}}$ \\
813: \hline
814: \hline
815:  Data  & 0.915 $\pm$  0.023 $\pm$ 0.021  \\
816: \hline
817: \hline
818: PYTHIA no-CR   & 0.918 $\pm$  0.003 \\
819: \hline
820:  SKI ($k_{\rm{I}}$=0.6)  & 0.896 $\pm$  0.003 \\
821: \hline
822:  SKI ($k_{\rm{I}}$=3.0)  & 0.843 $\pm$  0.003 \\
823: \hline
824:  SKI 100\% & 0.762 $\pm$  0.003  \\
825: \hline
826:  SKII   & 0.916 $\pm$  0.003  \\
827: \hline
828:  ARIADNE no-CR   & 0.929 $\pm$  0.003 \\
829: \hline
830:  AR2   & 0.919 $\pm$  0.003  \\
831: \hline
832:  HERWIG no-CR   & 0.948 $\pm$  0.005 \\
833: \hline
834:  HERWIG CR  & 0.946 $\pm$  0.005  \\
835: \hline
836: \end{tabular}
837: \caption[]{Measured value of $R_{\rm{N}}$ and model predictions.}
838: \label{tab:model}
839: \end{center}
840: \end{table}
841: 
842: 
843: 
844: \clearpage
845: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
846: %  FIGURES
847: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
848: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
849: 
850: \begin{figure}[htbp]
851: \begin{center}
852:     \includegraphics*[width=14cm]{fig1_paper.eps}
853: \end{center}
854: \caption[]{Determination of the $\phi_{i}$ angle for the particle $i$.}
855: \label{fig:plane}
856: \end{figure}
857: \begin{figure}[htbp]
858: \begin{center}
859:     \includegraphics*[width=8.2cm]{fig2a_paper.eps}
860:     \includegraphics*[width=8.2cm]{fig2b_paper.eps}
861: \end{center}
862: \caption[]{a) Particle- and b) energy-flow distributions 
863: at $\sqrt{s} = 189-209 \GeV$ for data and MC predictions.}
864: \label{fig:flow2}
865: \end{figure}          
866: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
867: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
868: \begin{figure}[htbp]
869: \begin{center}
870:     \includegraphics*[width=12.5cm]{fig3_paper.eps}
871: \end{center}
872: \vspace*{-1cm}
873: \caption[]{a) Particle-flow distribution as a function of the rescaled angle
874: for data and for PYTHIA MC  predictions  without CR,  and
875: with the SKI 100\% model.  Distributions of b)  combined intra-W particle flow and
876: c) combined inter-W particle flow.}
877: \label{fig:flow3}
878: \end{figure}          
879: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
880: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
881: \begin{figure}[htbp]
882: \begin{center}
883:     \includegraphics*[width=8.cm]{fig4a_paper.eps}
884:     \includegraphics*[width=8.cm]{fig4b_paper.eps}
885: \end{center}
886: \caption[]{ Ratio  of a) particle- and b) energy-flow distributions (Equation 4) in regions A+B 
887: to that in
888:  regions C+D. Statistical uncertainties are shown.
889: }
890: \label{fig:flowrat}
891: \end{figure}          
892: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
893: 
894: \begin{figure}[htbp]
895: \begin{center}
896: %    \includegraphics*[width=12.cm]{rnrootsyst_col.eps}
897:     \includegraphics*[width=12.cm]{fig5_paper.eps}
898: \end{center}
899: \caption[]{The ratio $R_{\rm{N}}$ as a function of $\rs$ at detector level
900:  for data and  PYTHIA no-CR and  SKI model predictions.
901: The parametrisation of the energy dependence is obtained by  fitting  a second order
902:  polynomial function to
903: the predicted MC dependence.
904: The parametrisation  obtained with PYTHIA no-CR gives
905: $R_{\rm{N}}(\rs)/R_{\rm{N}}$(189 {\rm{GeV}}) = $-3.07 \times 10^{-4} s + 0.1297 \rs -12.56$. 
906: The dependence obtained with the SKI model ($k_{\rm{I}}$= 3) leads to a 
907: 2.3\% change in the average rescaled $R_{\rm{N}}$ value at 189 \GeV.
908: }
909:  
910: \label{fig:ratio}
911: \end{figure}          
912: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
913: \begin{figure}[htbp]
914: \begin{center}
915:     \includegraphics*[width=8.2cm]{fig6a_paper.eps}
916:     \includegraphics*[width=8.2cm]{fig6b_paper.eps}
917: \end{center}
918: \vspace*{-1.0cm}
919: \caption[]{Particle-flow distributions a) before and b) after angle rescaling 
920: for the semileptonic W decays for 
921: data and KORALW prediction.}
922: \label{fig:flowlept}
923: \end{figure}          
924: 
925: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
926: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
927: \begin{figure}[htbp]
928: \begin{center}
929:     \includegraphics*[width=8.cm]{fig7_paper.eps}
930: \end{center}
931: \vspace*{-1.0cm}
932: \caption[]{a) Particle-flow distributions as a function of the rescaled angle
933: for the semileptonic W decays for 
934: data and the KORALW prediction.
935: b) Ratio of data and MC as a function of the rescaled angle. c) 
936: Ratio $R$  of the particle flow in hadronic events
937:  divided by twice the particle flow in semileptonic events.}
938: \label{fig:flowlept2}
939: \end{figure}          
940: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
941: \end{document}
942: 
943: 
944: