astro-ph0606415/ms.tex
1: %\documentclass{aastex}
2: 
3: 
4: %\documentclass[preprint]{aastex}
5: 
6: %% preprint2 produces a double-column, single-spaced document:
7: 
8: %\documentclass[preprint2]{aastex}
9: 
10: %\documentclass[12pt,preprint]{aastex}
11: %\usepackage{epsfig}
12: 
13: \documentclass{emulateapj}
14: %\usepackage{emulateapj5}
15: 
16: 
17: \bibliographystyle{apj}
18: 
19: 
20: \newcommand{\PSbox}[3]{\mbox{\rule{0in}{#3}\special{psfile=#1}\hspace{#2}}}
21: \newcommand{\msun}{{\rm M}_{\solar}}
22: \newcommand{\Msun}{{\rm M}_{\solar}}
23: \newcommand{\solar}{\ifmmode_{\mathord\odot}\;\else$_{\mathord\odot}\;$\fi}
24: \newcommand{\kms}{\, \rm{km}\,  \rm{s}^{-1}}
25: \newcommand{\kpc}{\, \rm{kpc}}
26: \def\ltsima{$\; \buildrel < \over \sim \;$}
27: \def\lsim{\lower.5ex\hbox{\ltsima}}
28: \def\Msunh{\mbox{$h^{-1}$M$_\odot$}}
29: \def\mpch{\mbox{$h^{-1}$Mpc}}
30: \def\Mpch{\mbox{$h^{-1}$Mpc}}
31: \def\rhovoid{\mbox{$\rho_{\rm void}$}}
32: \def\Mpc{{\rm Mpc}}
33: \def\W{{\rm W}}
34: \def\deg{\ifmmode{^\circ}\else{$^\circ$}\fi}
35: \def\hGpc{\ifmmode{h^{-1}{\rm Gpc}}\else{$h^{-1}{\rm Gpc}$}\fi}
36: \def\hkpc{\ifmmode{h^{-1}{\rm kpc}}\else{$h^{-1}{\rm kpc}$}\fi}
37: \def\hMpc{\ifmmode{h^{-1}{\rm Mpc}}\else{$h^{-1}{\rm Mpc}$}\fi}
38: \def\hMsun{\ifmmode{h^{-1}M_\odot}\else{$h^{-1}M_\odot$}\fi}
39: \def\km{{{\rm km}}}
40: \def\kms{{{\rm km}/{\rm s}}}
41: \def\kpc{{\rm kpc}}
42: \def\lyr{{\rm lyr}}
43: \def\muK{\ifmmode{\mu{\rm{K}}}\else{$\mu$K}\fi}
44: \def\mum{\ifmmode{\mu{\rm{m}}}\else{$\mu$m}\fi}
45: \def\m{{\rm m}}
46: \def\s{{\rm s}}
47: \def\pch{\mbox{$h^{-1}$pc}}
48: \def\sige{\mbox{$\sigma_8$}}
49: 
50: %\newcommand{\LCDM}{$\Lambda$CDM}
51: %\newcommand{\aj}{{AJ}}
52: %\newcommand{\apj}{{ApJ}}
53: %\newcommand{\apjl}{{ApJ}}
54: %\newcommand{\apjs}{{ApJS}}
55: %\newcommand{\mnras}{{MNRAS}}
56: %\newcommand{\nat}{{Nature}}
57: %\newcommand{\aap}{A\&A}
58: %\newcommand{\apss}{Ap\&SS}
59: %\newcommand{\araa}{ARA\&A}
60: %\newcommand{\prd}{Phys. Rev. D}
61: %\newcommand{\pra}{Phys. Rev. A}
62: %\newcommand{\pasj}{PASJ}
63: 
64: %\newcommand{\vdag}{(v)^\dagger}
65: %\newcommand{\myemail}{trujillo@mpia-de}
66: %\newcommand{\tgc}{\ensuremath{C_{r_e}(1/3)}}
67: 
68: %% You can insert a short comment on the title page using the command below.
69: 
70: %\slugcomment{ApJL, in press}
71: 
72: 
73: \shorttitle{The orientation of dark matter halos}
74: 
75: \shortauthors{Patiri et al.}
76: 
77: \begin{document}
78: 
79: 
80: \title{The alignment of dark matter halos with the cosmic web}
81: 
82: \author{Santiago G. Patiri\altaffilmark{1}, Antonio J. Cuesta\altaffilmark{2},
83:  Francisco Prada\altaffilmark{2}, Juan Betancort-Rijo\altaffilmark{1,3} and Anatoly Klypin\altaffilmark{4}} 
84: 
85: 
86: \altaffiltext{1}{Instituto de Astrof\'{\i}sica de Canarias, Calle V\'{\i}a
87: L\'actea, E-38200, La Laguna, Tenerife, Spain; spatiri@iac.es}
88: \altaffiltext{2}{Instituto de Astrof\'{\i}sica de Andaluc\'{\i}a (CSIC), E-18008, Granada, Spain; ajcv@iaa.es; fprada@iaa.es}
89: \altaffiltext{3}{Facultad de Fisica, Universidad de La Laguna, Astrof\'{\i}sico Francisco S\'anchez, s/n, La Laguna Tenerife E38200, Spain; jbetanco@iac.es}
90: \altaffiltext{4}{Astronomy Department, New Mexico State University, MSC 4500, P.O.
91: Box 30001, Las Cruces, NM 880003-8001; aklypin@nmsu.edu}
92: 
93: 
94: 
95: \begin{abstract}
96: 
97: We investigate the orientation of the axes and angular momentum of dark matter halos with 
98: respect to their neighboring voids using high resolution N-body cosmological simulations. 
99: We find that the minor axis of halos tends to be aligned along the line joining the halo with the center 
100: of the void, and that the major axis is perpendicular to this line. However, we find that the 
101: angular momentum of halos does not have any particular orientation.
102: These results may provide information about the mechanisms whereby the large-scale structure of the 
103: Universe affects galaxy formation and cast light upon the issue of the orientation of galaxy disks with 
104: respect to their host halos. 
105: 
106: 
107: \end{abstract}
108: \keywords{galaxies: halos  --- large--scale structure of the universe -- methods:
109: statistical --- dark matter}
110: 
111: 
112: 
113: \section{Introduction}
114: 
115: The knowledge of the intrinsic alignments of the angular momentum of dark matter halos 
116: with respect to their axes and the alignments of these axes with respect to the
117: galaxy disks is of great importance in order to understand the processes
118: involved in galaxy formation. Moreover, the alignment of halo axes, angular momentum and
119: galaxy spins with respect to the large-scale structure may provide further clues to this end; 
120: for example, they could be determinant to understand the origin of
121: the angular momentum of galaxies (see e.g. Navarro et al. 2004). They also may help to understand several
122: observational results like the distribution of galactic satellites 
123: (e.g. Zentner et al. 2005, Azzaro et al 2005), the formation of galactic warps (e.g. Dekel \& Shlosman 1983; Toomre
124: 1983; Bailin 2004) and the tidal streams observed for example in the Milky Way (e.g. Ibata et al. 2002). 
125: Furthermore, they could be a source of contamination in weak lensing studies (e.g. Heymans et al. 2006).
126: 
127: Dark matter halos are in general triaxial being defined by their
128: three semi-axes, $a \geq b \geq c$ (Dubinski \& Carlberg 1991; Warren et al. 1992; Cole \& Lacey 1996;
129: Jing \& Suto 2002; Allgood et al. 2006). Recently, some effort have been devoted to measure their 
130: intrinsic alignments with respect to the angular momentum. It has been found
131: that the angular momentum of dark matter halos tends to be parallel 
132: to their minor axis (Bailin \& Steinmetz 2005; Allgood et al 2006). 
133: This situation is typical for rotationally supported objects. However, as dark matter halos are 
134: found to be not rotationally supported, this alignment could have a non trivial origin (Allgood et al 2006).
135: 
136: On the other hand, the measurement of the alignments of galaxy disks with their host dark matter halos 
137: is both theoretical and observationally difficult. Using cosmological hydrodynamic simulations 
138: of disk galaxy formation, Bailin et al. (2005) found that the halo minor axis within 0.1$R_{vir}$ 
139: matches the disk axis. Beyond 0.1$R_{vir}$, the halo orientation is uncorrelated with the disk axis. 
140: Moreover, Hoekstra, Yee \& Gladders 2004, using galaxy-galaxy lensing measurements find that galaxies 
141: (disks plus ellipticals combined) are aligned with the major axis of their dark matter halo.
142: Other works show that the angular momentum of disks and dark matter are not perfectly aligned
143: (van den Bosch et al. 2002; Sharma \& Steinmetz 2005). Hence, the final orientation of the galactic 
144: disk in a cosmological dark matter halo is still an unresolved problem.
145: 
146: One way of gaining insight about these alignments is to study the alignments of both, halos and disks, with 
147: respect to the large-scale structure. The basic idea is that 
148: the angular momentum of halos and galaxies is generated by tidal torques
149: operating on the primordial material destined to form them (Peebles 1969; White 1984).
150: In this scenario, the primordial angular momentum of barions and dark matter are aligned and 
151: the final galactic angular momentum is a `relic' of the primordial one. According to this, 
152: the current misalignment between galactic and dark matter halo angular momentum is due to 
153: subsequent tidal torques affecting dark matter halos. However, this scenario could be more complex
154: because additionally to the primordial tidal torques operating on the material which 
155: eventually becomes the disk, the mass accretion by the disk is a clumpy and 
156: stochastic process (see Vitvitska et al. 2002).
157: Along this line, Navarro et al. (2004) found, using high resolution numerical simulations, that galaxy disks are
158: rather well aligned with the angular momentum of the surrounding matter at the epoch of turnaround.
159: Bailin \& Steinmetz (2005) found that the minor axes of halos of groups and clusters show a strong tendency to
160: lie perpendicular to the direction of filaments. However, it has been suggested that although filaments cause 
161: alignments of cluster-size halos, 
162: they have not discernible influence on smaller halos (Altay et al. 2006). On the other hand, it has recently been found using 
163: observations from the 2dFGRS and the SDSS, that galactic disk planes tend to be perpendicular to the walls 
164: of cosmic voids (Trujillo, Carretero \& Patiri 2006).
165: 
166: In this letter we study, using high resolution N-body simulations, the orientation of 
167: axes and angular momentum of galactic size dark matter halos embedded in the walls of 
168: large voids, those with radius larger than $10 \mpch$. We will discuss the 
169: observational implications of our results. 
170: 
171: %Through this work, we adopt a cosmological model with
172: %H$_0$=100 h km s$^{-1}$ Mpc$^{-1}$, $\Omega_m$=0.3 and $\Omega_\Lambda$=0.7.
173: 
174: 
175: \section{Numerical simulations and Methodology}
176: 
177: 
178: \subsection{N-body simulations}
179: 
180: We have used for our analysis a high resolution N-body numerical 
181: simulation of $512^{3}$ particles in a box of $120\mpch$ on a side assuming the
182: concordance $\Lambda$CDM cosmological model ($\Omega_0$ = 0.3,
183: $\Omega_\Lambda$ = 0.7 and $H_{0} = 100h \kms/\Mpc$ with $h = 0.7$). The 
184: simulation was done with the Adaptive Refinement Tree code (ART, Kravtsov et al. 1997). 
185: The force resolution was $1.8 \hkpc$ comoving and the mass of a particle $1.1 \times 10^{9} \mpch$.
186: Dark matter halos are identified in the simulation by the Bound Density Maxima 
187: algorithm (BDM, Klypin \& Holtzman 1997; Klypin et al. 1999). For more details about this 
188: simulation see Table 1 ($L120_{0.9}$) in Allgood et al. (2006). There is a total of 181000 halos 
189: in the whole simulation box at $z=0$. However, we limited our analysis to well-resolved halos
190: with more than 1000 particles inside $R_{vir}$ to be able to analyze accurately
191: their internal axes and angular momentum orientations.
192: 
193: \subsection{Searching for voids and halo sample}
194: 
195: In order to study the alignments of dark matter halos with their 
196: surrounding large-scale structure, we searched for large voids 
197: in the numerical simulation box using the HB void finder (Patiri et al. 2006).
198: The HB Void Finder is conceptually simple, as it searches
199: for the maximal spheres with radius larger
200: than a given value that are empty of halos with mass larger than
201: some specified value. This code is designed to be accurate
202: and computationally efficient in finding the largest
203: voids in any {\it point} distributions like galaxy samples or 
204: dark matter halos samples. To search for these voids, the code first
205: generates a sphere of fixed radius randomly located within
206: the sample. We check if this sphere is empty of
207: previous selected halos. If so, we increase the size of the sphere until
208: its surface reaches the four nearest halos. This is our void. For a
209: full description of the algorithm see Patiri et al. (2006).
210: 
211: In the present work, we have searched for voids defined by dark matter 
212: halos with mass larger than $1.0 \times 10^{12} \Msunh$. We choose this mass 
213: in order to compare with the observational results found by Trujillo, Carretero \& Patiri (2006),
214: i.e. we select the halo mass so that their number density is equal to the number density of 
215: galaxies found in the observed galaxy samples (galaxies brighter than $M_{b_{J}}=-19.4 +5log h$). 
216: With these definitions, we found a total of 60 voids with radius larger than $10\mpch$ in the whole simulation box.
217: Once we have obtained the voids, we selected the halos located within shells of 
218: $4 \mpch$ thick beyond the surface of the void. Also, as we are interested in the orientations 
219: of galactic size halos with respect to their neighbouring large-scale structure, we restricted the mass of the halos 
220: to be smaller than $1.0 \times 10^{13} \Msunh$. We detect a total of 1729
221: halos in this mass range that are located in the shells of large voids.
222: For each of these halos we obtain their principal axes ($\hat{a},\hat{b},\hat{c}$) and calculate their 
223: angular momentum, in order to measure their orientations with respect to the void centers (see next section).
224: 
225: 
226: \subsection{Measuring halo orientations}
227: 
228: We follow the method presented in Allgood et al.(2006) to fit our halos to ellipsoids. First of all, we 
229: calculate the \emph{reduced} inertial tensor for all the particles inside 
230: $R_{\mathrm{vir}}$ as $I_{ij}=\Sigma x_ix_j/r^2$, where the sum is over every particle inside a 
231: sphere of radius $R_{\mathrm{vir}}$ around the center of the halo. The halo center and 
232: $R_{\mathrm{vir}}$ are given by our BDM halo finder. The coordinates 
233: here are Cartesian and the denominator (squared distance to halo center) makes the eigenvalues 
234: of the tensor, $\lambda_1>\lambda_2>\lambda_3$ dimensionless and non-sensitive to particle distances. 
235: The axis lengths $a>b>c$ are proportional to the square root of the eigenvalues. After this first step, 
236: we iterate this procedure to get more and more approximate ellipsoids that fit our dark matter 
237: distribution. To achieve it, we take the eigenvalues that we have just calculated and we normalize them so 
238: that $a=R_{\mathrm{vir}}$ and the ratios $c/a$, $b/a$ are preserved. We then take into account every 
239: particle inside the ellipsoid (the ones that obey 
240: $\left(\frac{\vec{r}\cdot \hat{a}}{a}\right)^2 +\left(\frac{\vec{r}\cdot \hat{b}}{b}\right)^2 +\left(\frac{\vec{r}\cdot \hat{c}}{c}\right)^2 <1$, 
241: with $\hat{a},\hat{b},\hat{c}$ being the unit eigenvectors), and calculate the reduced inertial tensor, defined now as:
242: \begin{equation}
243: I_{ij}=\sum \frac{x_ix_j}{r_e^2} \qquad r_e=a\sqrt{x^2/a^2+y^2/b^2+z^2/c^2}
244: \end{equation}
245: 
246: Its diagonalization results in the new eigenvalues and axis lengths. This method is repeated until 
247: convergence is achieved in every axis, with a tolerance $\epsilon=5 \times 10^{-2}$. We stop 
248: the fit after just 7 iterations, as the convergence is usually very fast. The halos that could not be fitted to ellipsoids ($\sim 8$\%) are therefore
249: rejected.
250: 
251: Once the ellipsoid axes are measured, we calculate in a straightforward way the angle between the 
252: vector joining the center of the halo to the center of the void 
253: $\vec{R}=R\hat{R}\equiv\vec{R}_{\mathrm{halo}}-\vec{R}_{\mathrm{void}}$, and the principal halo axes with 
254: $\alpha=\arccos \vert\hat{R}\cdot\hat{a}\vert \quad \beta=\arccos \vert\hat{R}\cdot\hat{b}\vert \quad \gamma=\arccos \vert\hat{R}\cdot\hat{c}\vert$. 
255: The  $1 \sigma$ error for the direction of the axes is about $8^{\circ}$ (see equation $5$ in Bailin \& Steinmetz 2004).
256: 
257: As the particles inside the ellipsoid are known after the iterative method converges, we can calculate 
258: the total angular momentum vector of the halo as:
259: \begin{equation}
260: \vec{L}=\sum m\vec{r}\times\vec{v}
261: \end{equation}
262: 
263: where $\vec{r}$ is again the particle position referred to the center of the halo, and $\vec{v}$ is 
264: the particle velocity. Note that we are approximating this center with the real center of mass of 
265: the ellipsoid, otherwise there would be non-negligible additional terms to take into account. 
266: Then the angle $\theta$ between $\vec{L}$ and $\vec{R}$ is calculated as follows:
267: \begin{equation}
268: \theta=\arccos \big|{\frac{\vec{R}\cdot\vec{L}}{RL}}\big|
269: \end{equation}
270: 
271: There are several sources of noise when calculating these angles, so it is not wise to take every 
272: halo we have in our sample. In particular, we reject 
273: the halos whose axes are determined to within an uncertainty greater than $0.2~ \mathrm{rad}\simeq 11^{\circ}$. 
274: Moreover, we reject halos with degenerate axes, as the 
275: direction of these axes is ill-defined and arbitrary within a plane (see Jing \& Suto 2002). Therefore, we just 
276: take halos with axis ratios lower than 0.95, in order to avoid such degeneracy. After all, 
277: we have 1187 halos left for our analysis.
278: 
279: We also checked if our halos have the expected properties reported in the literature for their axis ratios and 
280: the intrinsic alignment of the angular momentum with the direction of principal axes. 
281: No discrepancies were found. The mean axis ratios in our mass range are 
282: $c/a=0.64\pm 0.14$, $b/a=0.78\pm 0.15$, and $c/b=0.82\pm 0.09$ in agreement with Jing \& Suto (2002), 
283: Bailin \& Steinmetz (2005), and Allgood et al. (2006). The median direction cosines of the angular 
284: momentum with the principal axes are $L\cdot a=0.32$, $L\cdot b=0.48$, and $L\cdot c=0.69$, in agreement 
285: with the results in Bailin \& Steinmetz (2005). About 75 \% of halos have an angle between their angular 
286: momentum and their major axes greater than $60^{\circ}$ as shown in Gottl\"ober \& Turchaninov (2006).
287: 
288: 
289: 
290: \section{Results and Conclusions}
291: 
292: 
293: In Figure 1 we show the orientations of the three halo axes and angular momentum 
294: with the cosmic voids. In the top-left panel we show the probability distribution 
295: of the $\alpha$ angle which is defined between the major axis of the dark matter halos 
296: and the vector pointing to the center of the void (black dots). Similarly, in the top-right and 
297: bottom-left panels we present the angles $\beta$ and $\gamma$ which are the ones 
298: between the middle and minor axis with the direction to the center of the void 
299: respectively. Finally, in the bottom-right panel we show the angle $\theta$ between 
300: the halo angular momentum and the direction to the center of the void. The solid line 
301: corresponds to the probability distribution of randomly distributed angles, which is a sine function for geometrical reasons.
302: 
303: In these plots we clearly see that halos are not randomly oriented within the shells of voids. 
304: The major axis of halos tends to lie along the plane perpendicular to the radial direction, 
305: i.e. there is an excess of halos with $\alpha$ angles near 90 degrees. On the contrary, 
306: the minor axis tend to point to the direction of the center of the void. There is a 
307: marginal orientation of the middle axes toward the plane perpendicular to the radial direction. 
308: However, we do not see any particular orientation of the halo angular momentum with respect to the
309: large-scale structure, confirming recent results obtained also from N-Body simulations using similar 
310: methodology (Heymans et al. 2006).  
311: 
312: To test the robustness of our results we have conducted several tests. First of 
313: all, we have analyzed the orientations of dark matter matter halos selected as 
314: described above, but in this case we used ``fake'' voids, i.e. we randomly 
315: placed spheres of fixed radius over the sample and calculated the angles between the 
316: axis (and angular momentum) of the dark matter halos located in the shells of these ``fake'' voids 
317: and the direction to their center. As expected, we found that the angles of the halos chosen 
318: to be in the shells of random spheres are uniformly distributed, i.e. they follow a sine distribution.
319: Moreover, we applied two different statistical tests: the average of the cosine test and a standard 
320: Kolmogorov-Smirnov test. If the angles are randomly distributed, the average of the cosine of these angles 
321: is $0.5$. So that, the bigger is the deviation from this value the larger will be the effect.
322: In Table 1 we show the results of the statistical tests. In the upper part of the table we show the results 
323: for the mean cosine (column 1) and the probabilities that the results be compatible with isotropy as given by 
324: the mean cosine and the KS tests (columns 2 and 3) for halos located in the shells of voids and, in the bottom 
325: part the same results but for halos inside the shells of the random spheres (i.e. the `fake' voids). From these results, 
326: we conclude that we can reject with high confidence (more than 7 sigmas) the null hypothesis (i.e. that the axis of dark 
327: matter halos are randomly distributed) for the halos in the shells of voids. As expected, the results 
328: of the statistical test for the halos located in the shells of the random spheres show that the 
329: orientation of these halos is compatible with the random distribution.
330: 
331: On the other hand, we used our data to gain some insight into the nature of the internal alignment of the 
332: minor axis with angular momentum found in previous works (e.g. Bailin \& Steinmetz 2004). We selected 
333: those halos where both directions form an angle smaller that $20\deg$. 
334: For these halos, we found that the alignment between the minor axis and the radial direction 
335: of the void is weaker than for randomly chosen halos, being even compatible with a random distribution.
336: 
337: Comparing these results with the observed tendency for the spin of spiral galaxies to point along 
338: the plane perpendicular to the direction of the center of the void found by Trujillo, Carretero \& Patiri (2006), three 
339: scenarios are possible: 1) The non-isotropy of the orientation of the spin of the galaxies is entirely 
340: induced by the halo alignments, 2) both alignments are two effects with common origin (namely, the 
341: tidal field in the shells of voids), or 3) a combination 
342: of the previous ones. In the first scenario the probability distribution for the angle between the major axis 
343: of the host halo and the spin of their galaxy is independent of the angle between the spin and the 
344: radial direction of the void. In this case, this probability distribution may be uniquely determined by 
345: comparing the distribution of the angle between the major axis and the radial direction with that of 
346: the angle between the spin of the galaxies and the radial direction. This would imply that the spin is very 
347: closely aligned with the major axis of the host halo. However, recent hydrodynamical simulations (Bailin et al. 2005) 
348: does not show this effect (they found that the spin of galaxies is uncorrelated with the halo). 
349: Therefore, the other two scenarios seem more plausible.  
350: 
351: 
352: \acknowledgments
353: 
354: 
355: We thank useful discussions with Conrado Carretero, Stefan Gottl\"ober, 
356: Andrey Kravtsov and Brant Robertson.
357: SGP, AJC, FP \& JBR thank support from the Spanish MEC under grant PNAYA 2005-07789. 
358: AJC acknowledges the financial support by the MEC through the spanish grant FPU AP2005-1826.
359: 
360: %\clearpage
361: 
362: \bibliography{ms.bib}
363: 
364: %\begin{thebibliography}{27}
365: %\expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
366: 
367: %\bibitem[Allgood et al.(2006)]{brandon} Allgood, B., Flores, 
368: %R.~A., Primack, J.~R., Kravtsov, A.~V., Wechsler, R.~H., Faltenbacher, A., 
369: %\& Bullock, J.~S.\ 2006, \mnras, 367, 1781 
370: 
371: %\bibitem[]{} Altay, G., Colberg, J. \& Croft, R. \ 2006, \mnras, 370, 1422
372: 
373: %\bibitem[]{} Azzaro, M., Zentner, A.R., Prada, F. \& Klypin, A. \ 2005, submitted to \apj, astro--ph/0506547
374: 
375: %\bibitem[Bailin(2004)]{2004PhDT} Bailin, J.\ 2004, Ph.D.~Thesis 
376: 
377: %\bibitem[Bailin \& Steinmetz(2004)]{BailinMatado2} Bailin, J., \& Steinmetz, M.\ 2004, \apj, 616, 27
378: 
379: %\bibitem[Bailin \& Steinmetz(2005)]{BailinMatado1} Bailin, J., \& Steinmetz, M.\ 2005, \apj, 627, 647 
380: 
381: %\bibitem[Bailin et al.(2005)]{bailinNew} Bailin, J., et al.\ 2005, \apjl, 627, L17 
382: 
383: %\bibitem[Cole \& Lacey(1996)]{colalacey} Cole, S., \& Lacey, C.\ 1996, \mnras, 281, 716 
384: 
385: %\bibitem[Dekel \& Shlosman(1983)]{1983IAUS..100..187D} Dekel, A., \& 
386: %Shlosman, I.\ 1983, IAU Symp.~100: Internal Kinematics and Dynamics of 
387: %Galaxies, 100, 187
388: 
389: %\bibitem[Dubinski \& Carlberg(1991)]{dubi} Dubinski, J., \& Carlberg, R.~G.\ 1991, \apj, 378, 496 %
390: 
391: %\bibitem[]{Stefan} Gottl\"ober S. \& Turchaninov V. \ 2006, EAS Publications Series, 20, 25
392: 
393: %\bibitem[]{}  Heymans, C., White, M., Heavens, A., Vale, C. \& Van Waerbeke, L. \ 2006, submitted to \mnras, astro--ph/0604001
394: 
395: %\bibitem[]{} Hoekstra H., Yee H.~K.~C. \& Gladders M.~D. \ 2004, \apj, 606, 67
396: 
397: %\bibitem[Ibata et al.(2002)]{Ibata} Ibata, R.~A., Lewis, G.~F., Irwin, M.~J., \& Quinn, T.\ 2002, \mnras, 332, 915 
398: 
399: %\bibitem[]{JS}  Jing Y. P. \& Suto Y. \ 2002, \apj, 574, 538
400: 
401: %\bibitem[Klypin \& Holtzman(1997)]{klypin} Klypin A. \& Holtzman J., 1997, astro--ph/9712217
402: 
403: %\bibitem[Klypin et al.(1999)]{KlypinSat1999} Klypin A., Kravtsov A.V.,
404: %  Valenzuela O., Prada F., 1999, \apj, 522, 82
405: 
406: %\bibitem[Kravtsov et al.(1997)]{1997ApJS..111...73K} Kravtsov, A.~V., 
407: %Klypin, A.~A., \& Khokhlov, A.~M.\ 1997, \apjs, 111, 73 
408: 
409: %\bibitem{nav04}  Navarro, J. F., Abadi, M. G. \& Steinmetz M., 2004, ApJ, 613, L41
410: 
411: %\bibitem{pat06}  Patiri S. G., Betancort-Rijo, J., Prada, F., Klypin, A. \&
412: %Gottl$\ddot{o}$ber, S., 2006, \mnras, 369, 335
413: 
414: %\bibitem{pee69}  Peebles, P.J.E., 1969, ApJ, 155, 393
415: 
416: %\bibitem[Sharma \& Steinmetz (2005)]{ss05} Sharma, S., \& Steinmetz, M. 2005, \apj, in press, astro--ph/0406533
417: 
418: %\bibitem[Toomre(1983)]{1983IAUS..100..177T} Toomre, A.\ 1983, IAU 
419: %Symp.~100: Internal Kinematics and Dynamics of Galaxies, 100, 177 
420: 
421: %\bibitem[Trujillo et al.(2006)]{bcp} Trujillo, I., Carretero, C., \& Patiri, S.~G.\ 2006, \apjl, 640, L111 
422: 
423: %\bibitem[van den Bosch et~al.(2002)]{vdb-etal02} van den Bosch, F.~C., Abel, T., Croft, R.~A.~C., Hernquist, L., \&
424: %  White, S.~D.~M. 2002, \apj, 576, 21
425: 
426: %\bibitem[Vitvitska et~al.(2002)]{vitvitska-etal02} Vitvitska, M., Klypin, A.~A., Kravtsov, A.~V., Wechsler, R.~H.,
427: %  Primack, J.~R., \& Bullock, J.~S. 2002, \apj, 581, 799%
428: 
429: %\bibitem[Warren et~al.(1992)]{warren-etal92} Warren, M.~S., Quinn, P.~J., Salmon, J.~K., \& Zurek, W.~H. 1992, \apj,
430: %  399, 405
431: 
432: %\bibitem{whi84}  White, S. D. M., 1984, ApJ, 286, 38
433: 
434: %\bibitem[Zentner et al.(2005)]{2005ApJ...629..219Z} Zentner, A.~R., 
435: %Kravtsov, A.~V., Gnedin, O.~Y., \& Klypin, A.~A.\ 2005, \apj, 629, 219 
436: 
437: %\end{thebibliography}
438: 
439: 
440: 
441: %\end{references}
442: \clearpage
443: 
444: \begin{deluxetable}{lccc}
445: \tablewidth{0pt}
446: \tablecaption{Results of the statistical tests}
447: \tablehead{ \colhead{Angle} &  \colhead{$<cos>$}   & \colhead{$<cos>$ Prob.}  & \colhead{KS Prob.}} 
448: \startdata
449: \cutinhead{ Halos in walls of voids}
450:  $\alpha$ & $0.459$  &  $1.13 \times 10^{-8} $  & $5.4 \times 10^{-7}$ \\
451:  $\beta$  & $0.481$  &  $7.58 \times 10^{-3} $  & $2.0 \times 10^{-3}$  \\
452:  $\gamma$ & $0.556$  &  $9.19 \times 10^{-15}$  & $9.4 \times 10^{-13}$  \\
453:  $\theta$ & $0.509$  &  $0.19$                  & $0.12$  \\
454: \cutinhead{ Halos in walls of random spheres }
455:  $\alpha$ & $0.496$  &  $0.48 $  & $0.23$ \\
456:  $\beta$  & $0.509$  &  $8.90 \times 10^{-2} $  & $7.80 \times 10^{-2}$  \\
457:  $\gamma$ & $0.492$  &  $0.19 $  & $0.10$  \\
458:  $\theta$ & $0.502$  &  $0.69 $  & $0.57$  \\
459: \enddata 
460: \end{deluxetable}
461: 
462: \clearpage
463: 
464: \begin{figure*}
465: %epsscale{1.}
466: %\epsfig{file=histon.eps,angle=0, width=\textwidth}
467: %\plotone{histon.eps}
468: %\vspace{-9cm}
469: \includegraphics[width=170mm]{f1.eps}
470: \caption{Probability density distribution for the angles between the axes (and angular momentum) of 
471: the dark matter halos and the vectors pointing to the center of the voids. 
472: Top-left panel: $\alpha$ angle for the major axis. Top-right panel: $\beta$ angle for the middle axis. 
473: Bottom-left panel: $\gamma$ angle for the minor axis. Bottom-right panel: $\theta$ angle for the angular
474: momentum. The error bar on each bin is the Poissonian error. The solid line represents randomly distributed
475: angles (i.e. a sine distribution)}
476: \label{fig:fig1} 
477: \end{figure*}
478: 
479: \end{document}
480: