ReadingTimeMachine/rtm-sgt-ocr-v1

1source,target2 The equation for p can beforunuulated as either an ordinary differeutial equation, The equation for $\rho$ can beformulated as either an ordinary differential equation3field above the active region a magnetofrictional relaxation techuique is applied.,field above the active region a magnetofrictional relaxation technique is applied.4 This technique evolves the coronal magnetic field through a sequence of related non-linear force-free equilibria., This technique evolves the coronal magnetic field through a sequence of related non-linear force-free equilibria.5 The technique is similar to that described in the vapors of vanBallegooijenetal.(2000) aud Mackay&vanDallegooijeu(2006a) where the code described iu these papers is adapted to a ocal cartesian frame of reference., The technique is similar to that described in the papers of \cite{bal00} and \cite{mac06a} where the code described in these papers is adapted to a local cartesian frame of reference.6 Previously the echnique has been successfully applied to consider he coronal field based on evolving plotospleric )oundarv conditions in either idealized setups (Mackay& Caizauskas20023:AlackayvanBallegooijon2005.2009) or approximations to observed masuetoegranis (Mackavetal.2000:Yeatesetal.2007. 2008).," Previously the technique has been successfully applied to consider the coronal field based on evolving photospheric boundary conditions in either idealized setups \citep{mac03,mac05,mac09} or approximations to observed magnetograms \citep{mac00,yea07,yea08}."7 Α key element of hese simulations is that through consideriug a -ue sequence of evolved equilibria. this allows he memory of previous flux connectivities aud currents to be maintained from oue tine to the rest as cherev and magnetic helicity is injected into the corona.," A key element of these simulations is that through considering a time sequence of evolved equilibria, this allows the memory of previous flux connectivities and currents to be maintained from one time to the next as energy and magnetic helicity is injected into the corona."8 Such a feature is significantly different compared to independent extrapolations which do not maintain such memory., Such a feature is significantly different compared to independent extrapolations which do not maintain such memory.9 The magnetic field. B=Vx.A. is evolved by the induction equation. where v(r.f) is the plasina velocity aud for the preseut study uon-ideal terms are uot included.," The magnetic field, ${\bf B} = \nabla \times {\bf A}$, is evolved by the induction equation, where ${\bf{v}}({\bf{r}},t)$ is the plasma velocity and for the present study non-ideal terms are not included."10 To cusure that the coronal field evolves through a series of force-free states a magneto-frictional method is cmploved (Yangαι 1986).," To ensure that the coronal field evolves through a series of force-free states a magneto-frictional method is employed (Yang, 1986)."11 We asstuue that the plasma velocity within the interior of the box Gepreseuting the solar corona) Is given by. where j=VosB aud v is the cocficieut of friction.," We assume that the plasma velocity within the interior of the box (representing the solar corona) is given by, where ${\bf{j}} = \nabla \times {\bf{B}}$ and $\nu$ is the coefficient of friction."12 Equation (3)) represents in au approximate manner. the fact that in the corona the Loreutz force is dominant and simulates the plasma," Equation \ref{eq:lor}) ) represents in an approximate manner, the fact that in the corona the Lorentz force is dominant and simulates the plasma"13"In (he past decade. a number of nearby star associations have been recognized as being quite voung. less (han 10Mvr old (e.g. IC 348. TW IIvcdrae. MDMI2., η Cha (Lada&Lada1999:Luliman&Steeghs 2004))).","In the past decade, a number of nearby star associations have been recognized as being quite young, less than $10\,\rm Myr$ old (e.g. IC 348, TW Hydrae, MBM12, $\eta$ Cha \citep{Lada95, Webb99, Luhman04}) )."14 Such associations are likely well stocked with recently formed. presumably bright eiant planets that shoukl in principle be easy prev for a varietv of planet detection technologies.," Such associations are likely well stocked with recently formed, presumably bright giant planets that should in principle be easy prey for a variety of planet detection technologies."15 Planning for the hunt. however. requires knowledge of the expected Iuminositv of voung giant. planets as a function of tme since (heir formation. particularly al voung ages when (μον are presumably easy. eae.," Planning for the hunt, however, requires knowledge of the expected luminosity of young giant planets as a function of time since their formation, particularly at young ages when they are presumably easy game."16 While models of the Iuminositv evolution of eiant. planets have a long pedigree (e.g..Graboskeetal. 1975).. the early work focused on the evolution of the solar svsteni giants. attempting to explain their current luminosity al an age of 4.5 Gyr.," While models of the luminosity evolution of giant planets have a long pedigree \citep[e.g.,][]{Grossman72, Graboske75}, the early work focused on the evolution of the solar system giants, attempting to explain their current luminosity at an age of 4.5 Gyr."17 Since planets lose memory of their initial conditions over time. initial conditions were selected more for computational convenience (han [or accuracy.," Since planets lose memory of their initial conditions over time, initial conditions were selected more for computational convenience than for accuracy."18 Many improvements Lave subsequently been made to the models. particularly in the characterization of jovian atmospheres at various effeclive temperatures. although essentially (he same initial conditions are still emploved (IIubbard:1980:Burrowsetal.1997:Chabrierοἱ2000).," Many improvements have subsequently been made to the models, particularly in the characterization of jovian atmospheres at various effective temperatures, although essentially the same initial conditions are still employed \citep{Hubbard80a, Burrows97, Chabrier00b}."19. The standard: evolution model begins with a hwdrogen-helium sphere having a large, The standard evolution model begins with a hydrogen-helium sphere having a large20ejivon in redshift coverage 2a1.5.,given in redshift coverage $2 < z < 4.5$.21 The lower bound in Nyy is due to the very rare meal detection in lower colum density svsteiis., The lower bound in $N_{\rm HI}$ is due to the very rare metal detection in lower column density systems.22 In. this range even if the line can be saturaed (depending on the Doppler width) Monte Carlo παΊος showed that fitting procedures of svutletic individual lines with simular resoution and S/N ratio of the observed spectra eive TT column density errors which are less han a few teus of dew (for b=25 logNyy=15.5. FWIIAL = 12 aud S/N = 2) this is tvpk‘ally 041 dew).," In this range even if the line can be saturated (depending on the Doppler width) Monte Carlo simulations showed that fitting procedures of synthetic individual lines with similar resolution and S/N ratio of the observed spectra give HI column density errors which are less than a few tens of $dex$ (for $b=25$, $\log N_{\rm HI} = 15.5$, FWHM = 12 and S/N = 20 this is typically 0.1 $dex$ )."23 The blending effect has a much more drnuafic 1upact on cohnun densitv uncertainties aud for this reason. we consider m the case of coiex structures as au mudividua cloud the total colmn densities of IIT aud of netal lines.," The blending effect has a much more dramatic impact on column density uncertainties and for this reason, we consider in the case of complex structures as an individual cloud the total column densities of HI and of metal lines."24 Estimating the heavy elemeif content in the clouds is urostly complicated by he poor knowledge of the ionising sources., Estimating the heavy element content in the clouds is mostly complicated by the poor knowledge of the ionising sources.25 As a firs Sa.uplification. we ASKme that this is dominated by photoionisation of the 1Nu vackeround aud negect any other mecha.," As a first simplification, we assume that this is dominated by photoionisation of the UV background and neglect any other mechanism."26" Collisioual ionisation is inipo""aut when the gas temperature execeds 107 1, At that teiiperatine, the Do»pler parameter for Is ll.. well above the mean value typically found 1iuLvo clouds."," Collisional ionisation is important when the gas temperature exceeds $10^5$ K. At that temperature, the Doppler parameter for HI is 41, well above the mean value typically found in clouds."27 The anaysis of meta lines 11i clouds (Rauch et al..," The analysis of metal lines in clouds (Rauch et al.,"28" 1997) slww that the mean “Doppler” teuerature iu these clouds is —L«104 I. masine any evidence of collisional ionisation hard to justify,"," 1997) shows that the mean “Doppler” temperature in these clouds is $\sim4\times10^4$ K, making any evidence of collisional ionisation hard to justify."29 Ouce the pliotoiotUsation equilibrium Is asstumed. we first cosider the subsample of clouS which show both CTV and SilV absorpticn.," Once the photoionisation equilibrium is assumed, we first consider the subsample of clouds which show both CIV and SiIV absorption."30 To calculate the metallicity wetse CLOUDY a assulue SIN cifferent shapes for the UV. background. normaized to the value at the Lyiia1 limit (Js25.107 cre s cm7 bx 1) changing f paralcter Sp=JiofJoos in the range 200JOO.," To calculate the metallicity we use CLOUDY and assume six different shapes for the UV background normalized to the value at the Lyman limit $J_{912} = 5\times10^{-22}$ erg $^{-1}$ $^{-2}$ $^{-1}$ $^{-1}$ ) changing the parameter $S_L =31J_{912}/J_{228}$ in the range $200-3000$."32" We varied. tje [C/T], a eas clensity in stch a way to reproduce the observed CIV.", We varied the [C/H] and gas density in such a way to reproduce the observed CIV.33 We alsο assunie t relative silicoutocarbon abundance to be beWe012: 0 and three tiues solar a consider the clotd size along the lune of sight to oin the range 1 spe <Rο kpc., We also assume the relative silicon–to–carbon abundance to be between 0 and three times solar and consider the cloud size along the line of sight to be in the range 1 kpc $\lsim R \lsim 50$ kpc.34 Given fiesο Πρες we obtain for lüs subsample a set of Ls [C/IT lueasurenients sτοσα in Fie. l..," Given these assumptions, we obtain for this subsample a set of 18 [C/H] measurements shown in Fig. \ref{f0}."35 Carbon alo11ance m clouds with detecte carbon and silicon has a laree spread with mea1 values of [C/TI 1.8 ane, Carbon abundance in clouds with detected carbon and silicon has a large spread with mean values of [C/H] $= -1.8$ and36To date. only three GCs have been surveved for correlations between Li aud proton-capture clements.,"To date, only three GCs have been surveyed for correlations between Li and proton-capture elements."37 Pasquini ct al. (, Pasquini et al. (382005) obtained Li abundances in niue TO members of NGC 6752: they found a depletion reaching down to —1 dex below the Spite plateau values. clearly iticorrelated with Na abundauces.,"2005) obtained Li abundances in nine TO members of NGC 6752: they found a depletion reaching down to $\sim$ 1 dex below the Spite plateau values, clearly anticorrelated with Na abundances."39 A similar result was tained by Bonifacio et al. (, A similar result was obtained by Bonifacio et al. (402007) who found a scatter in Li abundances much larger than observational errors and anticorrelated with Na amoue (oulv) four stars iu 17 Tuc.,2007) who found a scatter in Li abundances much larger than observational errors and anticorrelated with Na among (only) four stars in 47 Tuc.41 Very recently. Lind et al. (," Very recently, Lind et al. ("422009). targeting about LOO inaiu-sequenuce (MS) aud carly subgiaut branch stars in NGC 6397. detected for the first time a siguificaut anticorrelation between Li and Na in this CC.,"2009), targeting about 100 main-sequence (MS) and early subgiant branch stars in NGC 6397, detected for the first time a significant anticorrelation between Li and Na in this GC."43 This last investigation supersedes an older research by Bonifacio et al. (, This last investigation supersedes an older research by Bonifacio et al. (442002) based on only a few stars. aud hielliehtiug the iuportance of large samples of stars for similar studies.,"2002) based on only a few stars, and highlighting the importance of large samples of stars for similar studies."45 Iu this Letter we present Na. O. aud Li abundance determinations in ~90 unevolved TO stars of the GC lr Tucanae. providing theleryest database of this ine available in the literature so far.," In this Letter we present Na, O, and Li abundance determinations in $\sim$ 90 unevolved TO stars of the GC 47 Tucanae, providing the database of this kind available in the literature so far."46 We retrieved from the ESO Archive FLAMES-Ciratfte spectra (MEDUSA configuration) of 109 unevolved members of the metalrich GC lf Tuc (Program/ 081.D-02587: PI Z. Shen)., We retrieved from the ESO Archive FLAMES-Giraffe spectra (MEDUSA configuration) of 109 unevolved members of the metal-rich GC 47 Tuc (Program 081.D-0287; PI Z. Shen).47 The observations were carried out in Service mode in 2008 AueustSeptember: eratings HR15u CR —17.000). IIR1S (Ro 18.100). and WR20A GR— 16.000) were used to cover Τὰ (6707 D. OL(TTT. 7773. τπτ. Yo and Na (8182. μμ...5191 ) features.," The observations were carried out in Service mode in 2008 August–September; gratings HR15n $R\sim$ 17,000), HR18 $R\sim$ 18,400), and HR20A $R\sim$ 16,000) were used to cover Li (6707 ), O (7771, 7773, 7774 ), and Na (8183, 8194 ) features, respectively."48 The data reduction was performed with the ESO Ciüraffe usingCascaxo and following he standard procedure: bias subtraction. flat-field correction. wavelength calibration. aud optimal spectruni extraction.," The data reduction was performed with the ESO Giraffe using and following the standard procedure: bias subtraction, flat-field correction, wavelength calibration, and optimal spectrum extraction."49 A —careful sky subtraction— has je done using the taskσος iu order o take iuto account wavelength shifts aud possible differences iu cluission Line streneths between stellar and sky spectra., A careful sky subtraction has been done using the task in order to take into account wavelength shifts and possible differences in emission line strengths between stellar and sky spectra.50 It is noteworthy that the presence of he moon at more than ilbuninatioun along with diffuse clouds caused some of the observed frames to © severely contaminated and were thus discarded (the üeht of Áuesust L7th)., It is noteworthy that the presence of the moon at more than illumination along with diffuse clouds caused some of the observed frames to be severely contaminated and were thus discarded (the night of August 17th).51 Different exposures were then co-added to improve sigual-to-noise ratios. obtaining vpical values of ~50LOO per pixel (ucar the Li region at ~G7UR )).," Different exposures were then co-added to improve signal-to-noise ratios, obtaining typical values of $\sim$ 50–100 per pixel (near the Li region at $\sim$ 6708 )."52 The abundancecode analysis was carried out using the (developed by R. Cratton) and Ίππο» (1993) model atinospheres: for Na aud O. we derived abuudances from the equivalent widths (EVs). applying corrections due to uou-LTE effects (Gratton ot al.," The abundance analysis was carried out using the code (developed by R. Gratton) and Kurucz (1993) model atmospheres; for Na and O, we derived abundances from the equivalent widths (EWs), applying corrections due to non-LTE effects (Gratton et al."53 1999)., 1999).54 LTE Li abundances were instead obtained through spectral svuthesis., LTE Li abundances were instead obtained through spectral synthesis.55 Stellar parameters were derived as follows. (, Stellar parameters were derived as follows. (561) We obtained initial effective temperatures (Dag)) from the de-reddened (£(2pnr Tarris 1996) BV and the calibration |o» Alouso et al. (,1) We obtained initial effective temperatures ) from the de-reddened $E(B-V)$ =0.04; Harris 1996) $B-V$ and the calibration by Alonso et al. (571996).,1996).58 The adopted values were derived from a relation between V imaguitude aud color-basedTig., The adopted values were derived from a relation between $V$ magnitude and color-based.59. This reduces errors for individual stars. (, This reduces errors for individual stars. (602) The surface eravitics were estimated from the fundamental relation between AJ. EL. aud (logg=Lill|log(Arar.)los(L/L.)|LlogTog15.0117). bv adopting the bolometric correctious Alonso ct al.,"2) The surface gravities were estimated from the fundamental relation between $M$, $L$, and $\log g=4.44+ \log(M/M_\odot)-\log(L/L_\odot)+614\log T_{\rm eff} - 15.0447$ ), by adopting the bolometric corrections by Alonso et al."62 along with a distance modulus of 13:31 (IEaris 1996). while AZpo was calculated iu the same fashion described iu Cratton ot . (," along with a distance modulus of 13.34 (Harris 1996), while $M_{\rm TO}$ was calculated in the same fashion described in Gratton et al. ("632010).,2010).64 Finally we assumed for all our sample stars iuieroturbuleuce/ £21.00 kin 1 (note that none of the eleiieuts under scrutiny is stronely affected by © values)., Finally we assumed for all our sample stars microturbulence $\xi$ =1.00 km $^{-1}$ (note that none of the elements under scrutiny is strongly affected by $\xi$ values).65 The |N/Fe| ratios were calculated assuning as solar references log(Na) =6.21 aud log0(O) 8.79. values obtained m a fully consistent fashion from the analysis of the solar spectrum by Gratton ct al. (," The [X/Fe] ratios were calculated assuming as solar references $\log~n{\rm (Na)}_{\odot}$ =6.21 and $\log~n{\rm (O)}_{\odot}$ =8.79, values obtained in a fully consistent fashion from the analysis of the solar spectrum by Gratton et al. ("662003). aud 0.76 (Carretta et al.,"2003), and $-$ 0.76 (Carretta et al."67 20094) as cluster metallicity., 2009a) as cluster metallicity.68[Fe/UJ= Both internal (random) and external (svetematic) errors inav affect our analysis., Both internal (random) and external (systematic) errors may affect our analysis.69 A discussion of the first kind of uucertantv. which is predominant for Investigating star-to-star variations. can be found in Carretta et al. (," A discussion of the first kind of uncertainty, which is predominant for investigating star-to-star variations, can be found in Carretta et al. ("702007): here we just mention that errors in EWs are dominant. while the iuterual uncertainties ou stellar parameters (Tig)) cau be neglected (note that the distance of a star ou the colormaeuitucde diagram from the MS. main loci represents the major error source in the adopted temperature values).,"2007); here we just mention that errors in EWs are dominant, while the internal uncertainties on stellar parameters ) can be neglected (note that the distance of a star on the color-magnitude diagram from the MS main loci represents the major error source in the adopted temperature values)."71 Systematic errors due to the adopted set of model atimosphercs and/or to reddening aud Tg scale cannot be larger than ~0.1 dex (see Bonifacio et al., Systematic errors due to the adopted set of model atmospheres and/or to reddening and $_{\rm eff}$ scale cannot be larger than $\sim$ 0.1 dex (see Bonifacio et al.72 2002. 2007x aud Caatton et al.," 2002, 2007 and Gratton et al."73 2001): in particular. for Li. the combination of three-dinieusional aud non-LTE effects vields uceligible corrections (sec. 6g... Axplund et al.," 2001); in particular, for Li, the combination of three-dimensional and non-LTE effects yields negligible corrections (see, e.g., Asplund et al."74 2003)., 2003).75 All the abundance values with the corresponding nucertaimtics. along with magnitudes aud coordinates. are available oulv iu electronic version.," All the abundance values with the corresponding uncertainties, along with magnitudes and coordinates, are available only in electronic version."76 As shown iu Figure L.. the unevolved cluster stars reveal a very clear ο auticorrelation. consistently with what las been obtained for eiut members (Carretta et al.," As shown in Figure \ref{f:nao}, the unevolved cluster stars reveal a very clear $-$ O anticorrelation, consistently with what has been obtained for giant members (Carretta et al."77 2009b)., 2009b).78 We detected an oxveen variation of about ~0.80 ddex aud a change in Na of ~0.77 ddex. to vo Compared with 1 aud 0.82 (O and Na respectively) or @lants in Carretta et al. CELL(," We detected an oxygen variation of about $\sim$ dex and a change in Na of $\sim$ dex, to be compared with $\sim$ 1 and 0.82 (O and Na respectively) for giants in Carretta et al. ("792009h).,2009b).80" The interquartile ranges IQR|Na/O| are and £702-0.0LT. respectively,"," The interquartile ranges IQR[Na/O] are $\pm$ 0.067 and $\pm$ 0.041, respectively."81 These two estinates are consistent at lo evel: however since the dsvarfs are a more homogeneous sample (i.e. very similar atmospheric parauicters) and he spectral features used to measure a given element are iot the same in dwarfs and giauts. the latter being weaker and plagued bv higher uncertainties. both the spread (auaxinuniminin~) auc the IQR result sheltly higher or giants.," These two estimates are consistent at $\sigma$ level; however since the dwarfs are a more homogeneous sample (i.e., very similar atmospheric parameters) and the spectral features used to measure a given element are not the same in dwarfs and giants, the latter being weaker and plagued by higher uncertainties, both the spread (maximum–minimum) and the IQR result slightly higher for giants."82 It is noteworthy that we found au offset in Na, It is noteworthy that we found an offset in Na83spectrograph at (he William Ierschel Telescope on La Palma.,spectrograph at the William Herschel Telescope on La Palma.84 We cross-correlated (he spectra wilh a svnthetic template of the Ca triplet lines al ~850 nm (ο measure velocities 2002): the median velocity error was 5 kinsJ|. less than the UMis 8.8 kms! central velocity dispersion.," We cross-correlated the spectra with a synthetic template of the Ca triplet lines at $\sim 850$ nm to measure velocities \cite[e.g.][]{kleyna02}; the median velocity error was 5 $\rm km\,s^{-1}$, less than the UMi's 8.8 $\rm km\,s^{-1}$ central velocity dispersion."85 As our stars were mostly at large projected radii in UAL. we combined our data with the more centrally concentrated sample of Armandroll οἱ al. (," As our stars were mostly at large projected radii in UMi, we combined our data with the more centrally concentrated sample of Armandroff et al. ("861995). alter subtracting the mean velocity difference of the two data sets.,"1995), after subtracting the mean velocity difference of the two data sets."87 The combined data set contains 134 stars., The combined data set contains 134 stars.88 Because the observing run was curtailed by bad weather. the data cid nol extend (o a sufficiently large radius to allow a fit to the halo shape and anisotropy. (c.f.2002).," Because the observing run was curtailed by bad weather, the data did not extend to a sufficiently large radius to allow a fit to the halo shape and anisotropy \citep[c.f.][]{wilkinson02, kleyna02}."89. However. we noted that a histogram of stellar velocities near the clump on the northeast side of UMi's major axis appeared narrower than the &.&kms| Gaussian describing UMi's overall line of sight velocity distribution (Figure 1).," However, we noted that a histogram of stellar velocities near the clump on the northeast side of UMi's major axis appeared narrower than the $8.8\,\rm km\,s^{-1}$ Gaussian describing UMi's overall line of sight velocity distribution (Figure 1)."90 This observation suggested constructing a model of UM's population consisting of the sum of (wo Gaussians. one representing the underlving oy=S.8kms! Gaussian. and the other representing a sub-population of unknown traction /. velocity offset. ὃς. and velocity dispersion σ..," This observation suggested constructing a model of UMi's population consisting of the sum of two Gaussians, one representing the underlying $\sigma_0=8.8\,\rm km\,s^{-1}$ Gaussian, and the other representing a sub-population of unknown fraction $f$, velocity offset $v_s$, and velocity dispersion $\sigma_s$."91 Assuming an observational velocity uncertainty oy). the probability of observing a particular velocity in Chis moclel is and the likelihood of obtaining a ensemble of velocities fv;} with errors [05] is Next we scanned the face of UM in RA and Dee in 2’ increments.," Assuming an observational velocity uncertainty $\sigma_{\rm obs}$, the probability of observing a particular velocity in this model is and the likelihood of obtaining a ensemble of velocities $\{v_i\}$ with errors $\{{\sigma_{\rm obs}}_i\}$ is Next we scanned the face of UMi in RA and Dec in $2^\prime$ increments."92" At each point. we collected all velocities in a 6' radius aperture. and. computed the statistical likelihood of drawing these velocities from an 8.8 kms! Gaussian (Equation 2 with f= 0). as well as (he likelihood of drawing the velocities from each member of a grid of two-Guassian models (Equation 2. with f0.1.0.2...1.0. v,=—10.9...10. e,=0.5.1.0... 15.0)."," At each point, we collected all velocities in a $6^\prime$ radius aperture, and computed the statistical likelihood of drawing these velocities from an 8.8 $\rm km\,s^{-1}$ Gaussian (Equation 2 with $f=0$ ), as well as the likelihood of drawing the velocities from each member of a grid of two-Guassian models (Equation 2, with $f=0.1, 0.2 \ldots 1.0$, $v_s=-10, -9 \dots 10$, $\sigma_s=0.5,1.0\ldots 15.0$ )."93 When the likelihood of the best-fitting (wo-Gaussian model exceeded.| the likelihood of the sinele Gaussian model by a large factor. we deemed that aperture to contain a kinematic subpopulation distinct from the rest of UMis stars.," When the likelihood of the best-fitting two-Gaussian model exceeded the likelihood of the single Gaussian model by a large factor, we deemed that aperture to contain a kinematic subpopulation distinct from the rest of UMi's stars."94 Figure 2 shows the result of (he scanning procedure., Figure 2 shows the result of the scanning procedure.95 Large dots are apertures where we found a significant. (likelihood ratio > 107) sub-population., Large dots are apertures where we found a significant (likelihood ratio $>10^3$ ) sub-population.96" At the largest dot. the best two-Gaussian population {σι=0.5kins le,"," At the largest dot, the best two-Gaussian population ( $\sigma_s=\,0.5 \rm km\,s^{-1}$ ,"97pointed observations within the errors.,pointed observations within the errors.98 If we add dust to the model and fix Vy. to IN4. no successful fit can be achieved.," If we add dust to the model and fix $N_{\rm w}$ to $N_{\rm opt}$, no successful fit can be achieved."99 For example. fixing to the Galactic value. the fit still yields an unacceptably high - of about 11.," For example, fixing to the Galactic value, the fit still yields an unacceptably high $\chi^2_{\rm red}$ of about 11."100 Allowing for a steeper intrinsic spectrum. with D=2.9 as in the pointed observations. à dusty warm absorber with .Vy. = οι is consistent with the spectrum from the Survey observation.," Allowing for a steeper intrinsic spectrum, with $\Gamma = 2.9$ as in the pointed observations, a dusty warm absorber with $N_{\rm w}$ = $N_{\rm opt}$ is consistent with the spectrum from the Survey observation."101 However. the quality of the fit is slightly worse compared to the dust-free warm absorber.," However, the quality of the fit is slightly worse compared to the dust-free warm absorber."102 We conclude that. within the limits of low photon statistics. the Survey data are roughly consistent with the pointed observations.," We conclude that, within the limits of low photon statistics, the Survey data are roughly consistent with the pointed observations."103 The ASCA observation of wwas analyzed and discussed in detail by Brinkmann et al. (1996)), The ASCA observation of was analyzed and discussed in detail by Brinkmann et al. \cite{wpb96}) )104 and Brandt et al. (1997))., and Brandt et al. \cite{brandt97}) ).105 Therefore we describe in the following only the differences to the above papers in terms of data preparation and then turn to our new spectral analysis., Therefore we describe in the following only the differences to the above papers in terms of data preparation and then turn to our new spectral analysis.106" We used the ""Revision 2 data obtained from the ASCA public archive at Goddard Space Flight Center (GSEC) and applied the following conservative screening criteria: For the GIS the minimum elevation angle above the Earth's limb (ELV) was chosen to be 5°.", We used the 'Revision 2' data obtained from the ASCA public archive at Goddard Space Flight Center (GSFC) and applied the following conservative screening criteria: For the GIS the minimum elevation angle above the Earth's limb ) was chosen to be $5\degr$ .107 In the case of the SIS we used ELV-107., In the case of the SIS we used $ >10\degr$.108 To avoid atmospheric contamination. data were only accepted in the SIS when the angle between the target and the bright earth (ER_EARTH) was greater than 207.," To avoid atmospheric contamination, data were only accepted in the SIS when the angle between the target and the bright earth ) was greater than $20\degr$."109 A minimum cut-off rigidity (COR) of 6 GeV/e was applied for both. the SIS and the GIS.," A minimum cut-off rigidity ) of 6 GeV/c was applied for both, the SIS and the GIS."110 Data taken within four read-out cycles of the CCD detectors after the passage of the South Atlantic Anomaly (SAA) and the day-night terminator are not considered in the analysis., Data taken within four read-out cycles of the CCD detectors after the passage of the South Atlantic Anomaly (SAA) and the day-night terminator are not considered in the analysis.111 Further. periods of high background were manually excluded from the data by checking the light curve of the observation.," Further, periods of high background were manually excluded from the data by checking the light curve of the observation."112 Source counts were extracted from a circular region centered on the target with a radius of 6 for the GIS and 1’ for the SIS., Source counts were extracted from a circular region centered on the target with a radius of $6\arcmin$ for the GIS and $4\arcmin$ for the SIS.113 We used the local background determined from the observation in the analysis for both detectors., We used the local background determined from the observation in the analysis for both detectors.114 In particular. the GIS background was estimated from asource free region at the," In particular, the GIS background was estimated from asource free region at the"115the Galactic plane (Sunvaey et al;,"the Galactic plane (Sunyaev et al.,"116 1993) and detection of the 6.4 keV Duorescent line with  keV equivalent. width from Ser B2 strongcomplex and few other clouds in the Galactic1 Center region (Ixovama. 1994. 1996) that the neutral matter of these clouds is (or was) illuminated by suggest X-ray. radiation. which gave rise to the reprocessed radiation.," 1993) and detection of the strong 6.4 keV fluorescent line with $\sim$ 1 keV equivalent width from Sgr B2 complex and few other clouds in the Galactic Center region (Koyama, 1994, 1996) suggest that the neutral matter of these clouds is (or was) illuminated by powerful X-ray radiation, which gave rise to the reprocessed radiation."117 Lt is powerful to stress that the value of the 6.4 keV. line equivalent width importantdetected ASCA from the largeSer D2 cloud that we observe an almost. by component without direct suggestsemission from the source., It is important to stress that the large value of the 6.4 keV line equivalent width detected by ASCA from the Sgr B2 cloud suggests that we observe an almost pure reprocessed component without direct emission from the source.118 I is puretherefore reprocessed that. this source is now in a weak state., It is therefore likely that this source is now in a weak state.119 One of the interesting hypothesislikely that reprocessed radiation was produced: by a short. outhurst of the suggestsX-ray. emission. from Ser A* few hunelrecl vears ago., One of the interesting hypothesis suggests that reprocessed radiation was produced by a short outburst of the X-ray emission from Sgr A* few hundred years ago.120 Simple estimates show that Ser A* luminosity at the level of 107 eres is to power Duorescent cniission [rom the Ser Be cloud. (ic. well in the requiredrange allowed by the Exldington limit for this object)., Simple estimates show that Sgr A* luminosity at the level of $10^{39}$ erg/s is required to power fluorescent emission from the Sgr B2 cloud (i.e. well in the range allowed by the Eddington limit for this object).121 We consider below a few simple observational tests which could verily this , We consider below a few simple observational tests which could verify this hypothesis.122The simplest test hypothesis.(ANAL. NMM. ASTPRO-E) would be to look for time," The simplest test (AXAF, XMM, ASTRO-E) would be to look for time"123Chromospheric evaporation relers to the drastic mass motions in flaring loops caused by rapicl energy deposit in chromospheric lavers (or probably higher) by non-thermal electrons or by thermal conduction.,Chromospheric evaporation refers to the drastic mass motions in flaring loops caused by rapid energy deposit in chromospheric layers (or probably higher) by non-thermal electrons or by thermal conduction.124 When the flare enerev is transported toward lower lavers. it can produce a local overpressure (hat drives both upward and downward mass motions.," When the flare energy is transported toward lower layers, it can produce a local overpressure that drives both upward and downward mass motions."125 These motions can be detected (hrough Doppler shilt measurements in chromospheric and coronal lines.Antonueciοἱal.(1982. 1985)..Antonueci&Dennis (1983).. Cantieldetal.(1987).," These motions can be detected through Doppler shift measurements in chromospheric and coronal lines.\cite{anto82,anto85}, \cite{anto83}, , \cite{canf87},"126 These motions can be detected (hrough Doppler shilt measurements in chromospheric and coronal lines.Antonueciοἱal.(1982. 1985)..Antonueci&Dennis (1983).. Cantieldetal.(1987)..," These motions can be detected through Doppler shift measurements in chromospheric and coronal lines.\cite{anto82,anto85}, \cite{anto83}, , \cite{canf87},"127of both temperature and abundance.,of both temperature and abundance.128" One spectral fit, for region A, finds a significantly higher temperature than the map suggests."," One spectral fit, for region A, finds a significantly higher temperature than the map suggests."129" This appears to be a smoothing issue, the spectral extraction regions used to create the map being significantly larger than region A, which contains only ~900 net counts (0.7-7.0 keV)."," This appears to be a smoothing issue, the spectral extraction regions used to create the map being significantly larger than region A, which contains only $\sim$ 900 net counts (0.7-7.0 keV)."130" The abundance measurements typically have larger uncertainties, and a greater variation in map pixel values, particularly in the highest abundance regions."," The abundance measurements typically have larger uncertainties, and a greater variation in map pixel values, particularly in the highest abundance regions."131" The highest abundance of any spectral fit is found in region B. The region contains ~1200 net counts, again suggesting that smoothing lowers the maps values, but the spectral fit agrees with the map within the uncertainties."," The highest abundance of any spectral fit is found in region B. The region contains $\sim$ 1200 net counts, again suggesting that smoothing lowers the maps values, but the spectral fit agrees with the map within the uncertainties."132" The regions with lowest abundances (regions C,D,E,F and G) appear to have map values which are slightly overestimated."," The regions with lowest abundances (regions C,D,E,F and G) appear to have map values which are slightly overestimated."133" However, they are again consistent within the uncertainties."," However, they are again consistent within the uncertainties."134" The abundances measured in the map spectral fits could be biased by a number of issues which arise from the assumption of a simple single temperature plasma model when fitting spectra which must, on some level, be produced by gas whose properties vary within the spectral extraction region."," The abundances measured in the map spectral fits could be biased by a number of issues which arise from the assumption of a simple single temperature plasma model when fitting spectra which must, on some level, be produced by gas whose properties vary within the spectral extraction region."135" One possible problem is the inverse Fe-bias (???),, which can affect abundance measurements"," One possible problem is the inverse Fe-bias \citep{Rasiaetal08,Simionescuetal09,Gastaldelloetal10}, which can affect abundance measurements"136in the innermost 4 areminutes of Pal I4 selected along the cluster Red Giant Branch.,in the innermost 4 arcminutes of Pal 14 selected along the cluster Red Giant Branch.137 It has been constructed from high-resolution (45.000<R 60.000) spectra collected with two different spectrographs: UVES@VLT and HIRESCKecklI. A detailed description of the reduction. procedure and of the radial velocity measure can be found in 109.," It has been constructed from high-resolution $45,000<R<60,000$ ) spectra collected with two different spectrographs: UVESVLT and HIRESKeckI. A detailed description of the reduction procedure and of the radial velocity measure can be found in J09."138 The high-resolution allows to obtain very accurate radial velocities with errors of ~0.3funis., The high-resolution allows to obtain very accurate radial velocities with errors of $\sim0.3~km/s$.139 A number of outliers (field stars)can be easily identified at velocities Av=|v—v|>5kms., A number of outliers (field stars)can be easily identified at velocities $\Delta v\equiv\vert v-\overline{v}\vert>5~km/s$.140 The bona-fide cluster members turn out to constitute à sample of 16 stars plus one star (2115) lying at Av=2.35fun (~ 46) so that it is not clear if 1t is a true cluster member. à binary star or a field outlier.," The bona-fide cluster members turn out to constitute a sample of 16 stars plus one star 15) lying at $\Delta v=2.35~km/s$ $\sim 4\sigma$ ) so that it is not clear if it is a true cluster member, a binary star or a field outlier."141 In the following analysis we will always consider the two samples defined including and excluding this star., In the following analysis we will always consider the two samples defined including and excluding this star.142 The error weighted averages of the bona-fide cluster members are v=72.343:0.05Kun/s (without star #115) and v272.31£0.05Kns Cwith star 115)., The error weighted averages of the bona-fide cluster members are $\overline{v}=72.34\pm0.05~km/s$ (without star 15) and $\overline{v}=72.31\pm0.05~km/s$ (with star 15).143 To calculate the velocity dispersion we searched for the value of σι that maximizes the logarithm of the probability density where v; and 6; are the velocity of the (th star and its associated uncertainty (see Pryor Meylan 1993)., To calculate the velocity dispersion we searched for the value of $\sigma_{v}$ that maximizes the logarithm of the probability density where $v_{i}$ and $\delta_{i}$ are the velocity of the $i$ -th star and its associated uncertainty (see Pryor Meylan 1993).144" The so-calculated velocity dispersions turn out to be σι=0.3974Kms (without star #115) and a,=0.66715kin (with star #115).", The so-calculated velocity dispersions turn out to be $\sigma_{v}=0.39_{-0.09}^{+0.14}~km/s$ (without star 15) and $\sigma_{v}=0.66_{-0.12}^{+0.19}~km/s$ (with star 15).145 The models adopted in this paper are based on a set of N-body simulations performed in the framework of both the Newtonian and MOND theories of gravity considering. the cluster immersed in the gravitational field of the Milky Way., The models adopted in this paper are based on a set of N-body simulations performed in the framework of both the Newtonian and MOND theories of gravity considering the cluster immersed in the gravitational field of the Milky Way.146 As we will show in Sect. 3.2..," As we will show in Sect. \ref{nbody_sec},"147 the effect of the external field is an essential ingredient for a proper comparison of the velocity dispersion of this low-mass cluster with observations. in particular when the MOND theory is considered.," the effect of the external field is an essential ingredient for a proper comparison of the velocity dispersion of this low-mass cluster with observations, in particular when the MOND theory is considered."148 It is however instructive to show also the predictions of a set of isolated analytical models to provide the necessary reference to quantify the effect of the external field and to investigate the effect of different degrees of anisotropy., It is however instructive to show also the predictions of a set of isolated analytical models to provide the necessary reference to quantify the effect of the external field and to investigate the effect of different degrees of anisotropy.149 In the following sections we will describe the complete set of models and simulations used in this paper., In the following sections we will describe the complete set of models and simulations used in this paper.150" Under the hypothesis that the cluster is isolated and spherically symmetric. simple models can be constructed by solving the Jeans equation p is the density. P is the gravitational potential. ar) is the anisotropy parameter. and o, and σι are. respectively. the radial and tangential components of the velocity dispersion tensor at the radius +."," Under the hypothesis that the cluster is isolated and spherically symmetric, simple models can be constructed by solving the Jeans equation where $\rho$ is the density, $\Phi$ is the gravitational potential, is the anisotropy parameter, and $\sigma_{r}$ and $\sigma_{t}$ are, respectively, the radial and tangential components of the velocity dispersion tensor at the radius $r$."151" If the system is self-gravitating. p and ® are related in Newtonian gravity by the Poisson equation while in MOND by the corresponding modified field equation (Bekenstein Milgrom 1984) where μία) 1s the so-called ""interpolating function"" such that μία)~x at x«I (the so-called ""deep-MOND"" regime) and μία)~| at xc9| (the Newtonian regime)."," If the system is self-gravitating, $\rho$ and $\Phi$ are related in Newtonian gravity by the Poisson equation while in MOND by the corresponding modified field equation (Bekenstein Milgrom 1984) where $\mu(x)$ is the so-called ""interpolating function"" such that $\mu(x)\sim x$ at $x\ll 1$ (the so-called ""deep-MOND"" regime) and $\mu(x)\sim1$ at $x\gg 1$ (the Newtonian regime)."152" In. the following we use the ""simple"" interpolating function j:(v)=ΥΓ) (Famaey Binney 2005)."," In the following we use the ""simple"" interpolating function $\mu(x)=x/(1+x)$ (Famaey Binney 2005)."153" For finite-mass M systems the boundary conditions of equations (4))vector, and (5)) are |V®]+0 >x. where r is the position "," For finite-mass isolated systems the boundary conditions of equations \ref{pois_eq}) ) and \ref{mond_eq}) ) are $|154\nabla\Phi |\to 0$ for $|{\bf r}|\to \infty$, where ${\bf r}$ is the position vector."155"For given density distribution and .(7). equation (2)) canbe solved to obtain σ,."," For given density distribution and $\beta(r)$, equation \ref{jeans_eq}) ) canbe solved to obtain $\sigma_{r}$."156 The tangential component of the velocity dispersion σι is then derived from eq. (3)), The tangential component of the velocity dispersion $\sigma_{t}$ is then derived from eq. \ref{beta_eq}) )157" and the line-of-sight (LOS) velocity dispersion at any given projected distance R from the center is given by where is the projected density at R ands, 1s the tidal radius.", and the line-of-sight (LOS) velocity dispersion at any given projected distance $R$ from the center is given by where is the projected density at $R$ and $r_t$ is the tidal radius.158 The above procedure allows to calculate a LOS velocity dispersion profile for any given choice of M/L and 03) according to both Newtonian and MOND theories., The above procedure allows to calculate a LOS velocity dispersion profile for any given choice of $M/L$ and $\beta(r)$ according to both Newtonian and MOND theories.159 We adopted the density profile of the King (1966) model that best-fits the data of SII defined by the parameters (Wo.ri.0)20.0.61/.25.04magarcsec7) which have been converted in physical units assuming a distance of d=71 kpe and two different values of the mass-to-light ratio: M/Ly=1.885 (derived for Pal 14. from stellar population synthesis by McLaughlin van der Marel 2005) and M/Ly=0.747 (corresponding to the minimum mass estimated by JO9 from star counts in the color-magnitude diagram).," We adopted the density profile of the King (1966) model that best-fits the data of S11 defined by the parameters $(W_{0},r_{c},\mu_{V,0})=(7,0.61\arcmin, 25.04~mag~arcsec^{-2})$ which have been converted in physical units assuming a distance of $d=71$ kpc and two different values of the mass-to-light ratio: $M/L_{V}$ =1.885 (derived for Pal 14 from stellar population synthesis by McLaughlin van der Marel 2005) and $M/L_{V}$ =0.747 (corresponding to the minimum mass estimated by J09 from star counts in the color-magnitude diagram)."160 Hereafter. we will refer to the M/L ratio in the V band simply as M/L. Regarding the degree of anisotropy. we considered. besides the isotropic case (.)2 0). two extreme anisotropic cases: purely tangential (32 —x) and maximally radial.," Hereafter, we will refer to the M/L ratio in the V band simply as M/L. Regarding the degree of anisotropy, we considered, besides the isotropic case $\beta=0$ ), two extreme anisotropic cases: purely tangential $\beta=-\infty$ ) and maximally radial."161" In the latter case we adopted the Osipkov-Merritt parametrization (Osipkov 1979; Merritt 1985) where 7, ts the anisotropy radius. which sets the boundary where orbits become significantly radially biased (systems with smaller 7, are more radially anisotropic)."," In the latter case we adopted the Osipkov-Merritt parametrization (Osipkov 1979; Merritt 1985) where ${r}_{a}$ is the anisotropy radius, which sets the boundary where orbits become significantly radially biased (systems with smaller ${r}_{a}$ are more radially anisotropic)."162 We assumed FaFuge Where Pymin IS the bona-fide minimum value for of ry for stability against radial-orbit instability.," We assumed ${r}_{a}={r}_{a,min}$, where ${r}_{a,min}$ is the bona-fide minimum value for of ${r}_{a}$ for stability against radial-orbit instability."163" In particular. we adopt 7,5,=2.8r. for the Newtonian model and ryjincondition=3.17. for the MOND one. corresponding to the marginal for stability 27,5,5/T,nap71.5 (where Trj, and T;4,5; are the radial and tangential kinetic energies computed within the half-mass radius ρω Nipoti et al."," In particular, we adopt ${r}_{a,min}=2.8 r_c$ for the Newtonian model and ${r}_{a,min}=3.1 r_c$ for the MOND one, corresponding to the marginal condition for stability $2T_{r,half}/T_{t,half}\simeq 1.5$ (where $T_{r,half}$ and $T_{t,half}$ are the radial and tangential kinetic energies computed within the half-mass radius $r_{half}$ ; Nipoti et al."164 2011; Ibata et al., 2011; Ibata et al.165 2011a)., 2011a).166 The LOS velocity dispersion profiles of all these models. caleulated by integrating numerically the above equations. are shown in Fig. l..," The LOS velocity dispersion profiles of all these models, calculated by integrating numerically the above equations, are shown in Fig. \ref{sig}. ."167 It is evident that. at least as long as the cluster is treated as isolated. MOND models predict a significantly larger velocity dispersion with respect to Newtonian ones (see also Baumgardt et al.," It is evident that, at least as long as the cluster is treated as isolated, MOND models predict a significantly larger velocity dispersion with respect to Newtonian ones (see also Baumgardt et al."168 2005: Sollima, 2005; Sollima169tto construct light curves in 5 different bands. 0.20.5. 1.12.25 and 510 keV. The time lags between the 0.5 keV soft light curve and all the harder light curves are shown in Fig. 7..,"to construct light curves in 5 different bands, 0.2–0.5, 0.5--1, 1–2, 2–5 and 5–10 keV. The time lags between the 0.2--0.5 keV soft light curve and all the harder light curves are shown in Fig. \ref{lags_vs_e}."170 Here the lags are plotted: as a function of the average-energv ratio between each hard. band. and the 0.20.5 band Gg Es). For the three lowest. Fourier frequeney bins. which had the smallest. error. bars.," Here the lags are plotted as a function of the average-energy ratio between each hard band and the 0.2–0.5 band $E_H/E_S$ ), for the three lowest Fourier frequency bins, which had the smallest error bars."171 In. all cases. the lags increase linearly. with the logarithm of the energy.," In all cases, the lags increase linearly with the logarithm of the energy."172 The best-fitting relations. plotted in solid lines in Fig. 7.. ," The best-fitting relations, plotted in solid lines in Fig. \ref{lags_vs_e}, ,"173"have functional forms: rife)=12331og"">Egfle.—196 for T.4.-10 Hz. z(E)2135logEgIs.125 [or 2.0.10+ Iz and rif)=199logμες38 [or 3.210? Lz."," have functional forms: $\tau(E)=1233 \log174E_H/E_S -196$ for $7.4\times10^{-5}$ Hz, $\tau(E)= 735 \log E_H/E_S175-176 125$ for $2.0\times10^{-4}$ Hz and $\tau(E)= 199 \log E_H/E_S -38$ for $3.2\times10^{-4}$ Hz."177 The errors in the lag values take into account the red-noise nature of the light curves ancl probably overestimates the error. between dillerent. energy. bands., The errors in the lag values take into account the red-noise nature of the light curves and probably overestimates the error between different energy bands.178 Therefore. he 47 values for the fits are not significant.," Therefore, the $\chi^2$ values for the fits are not significant."179 X. perfect og-linear relation between lag amplitude and energy. ratio should cross (1.0) in this plot. as the lags should tend to 0 when calculated between identical energy. bands.," A perfect log-linear relation between lag amplitude and energy ratio should cross (1,0) in this plot, as the lags should tend to 0 when calculated between identical energy bands."180 Therefore. he log-linear relation might not hold down to small energy dillerences.," Therefore, the log-linear relation might not hold down to small energy differences."181 This possible change in the energy dependence is not significant however. given that if we force the intercepts o equal 0. the fits are still acceptable.," This possible change in the energy dependence is not significant however, given that if we force the intercepts to equal 0, the fits are still acceptable."182 A similar linear dependence of the lags on the logarithim of the energy. ratio has been observed for the ΗΝ Cve δν (Cuietal.1997:Crary1998:Nowakct1999)... sugeesting that a similar mechanism operates in both sources to produce the time lags.," A similar linear dependence of the lags on the logarithm of the energy ratio has been observed for the BHXRB Cyg X-1 \citep{Cui,183 Crary,Nowak_lags}, suggesting that a similar mechanism operates in both sources to produce the time lags."184 Sienilicant time lags between X-rav energy bands have been measured for a [ον AGN. NGC 7469. (Papadakis.Nandra&Ixazanas 2001)... ALCG6-30-15 (Vaughanetal. 2003).. NGC 4051 (Mellardyctal.2004) and NGC 3783 (Markowitz2005).," Significant time lags between X-ray energy bands have been measured for a few AGN, NGC 7469 \citep{Papadakis_7469}, MCG–6-30-15 \citep{VaughanMCG}, NGC 4051 \citep{McHardy4051} and NGC 3783 \citep{Markowitz3783}."185. In all cases. the lags appear to increase with variability time-scale and. where it has been possible to measure. also with energy separation of the bands.," In all cases, the lags appear to increase with variability time-scale and, where it has been possible to measure, also with energy separation of the bands."186 Previous lag spectra were normally fitted with single power law models and. in many cases. only the amplitude of the lags could be Left as a free parameter. due to the quality of the data.," Previous lag spectra were normally fitted with single power law models and, in many cases, only the amplitude of the lags could be left as a free parameter, due to the quality of the data."187 Consequently. even if there were intrinsic slope changes in the lag spectra of any other AGN. they would not have been detected.," Consequently, even if there were intrinsic slope changes in the lag spectra of any other AGN, they would not have been detected."188 Jo allowa direct. comparison of the amplitude of the lags in dillerent objects. we calculated lag spectra for," To allowa direct comparison of the amplitude of the lags in different objects, we calculated lag spectra for"189The hardness of the radiation field also contributes to the variations in intensity of the various PAI bands.,The hardness of the radiation field also contributes to the variations in intensity of the various PAH bands.190 Laboratory preclictions (e.g..Allamanclolaοἱal.1999:Nim&Savkally2002:Dauschlieher2002) suggest that ionized PAIL molecules produce stronger CC mode vibrations (e.g..Bakesetal.," Laboratory predictions \citep[e.g.,][]{Allamandola99,Kim02,Bauschlicher02} suggest that ionized PAH molecules produce stronger CC mode vibrations \citep[e.g.,][]{Bakes01}."1912001).. Bauschlicheretal.(2008). finds that the intensities of the CC! modes arising from cationic PAIIs are increased by an order of magnitude or more compared to neutral species., \citet{Bauschlicher08} finds that the intensities of the CC modes arising from cationic PAHs are increased by an order of magnitude or more compared to neutral species.192 Observationally this manilests itself as à strengthening of the CC modes (ie.. the 6.2 jan. 7.7 jim and 8.6 yan PAIL features) relative to the CII modes (primarily the 11.3 pam feature. although the 8.6 jii feature is also affected to a lesser extent) in a given source.," Observationally this manifests itself as a strengthening of the CC modes (i.e., the 6.2 $\mu$ m, 7.7 $\mu$ m and 8.6 $\mu$ m PAH features) relative to the CH modes (primarily the 11.3 $\mu$ m feature, although the 8.6 $\mu$ m feature is also affected to a lesser extent) in a given source."193 Quantitative studies have been undertaken (o attempt (to connect these variations of PAIL intensities to the physical conditions of the gas in a variety of astrophysical environments (see further discussion below)., Quantitative studies have been undertaken to attempt to connect these variations of PAH intensities to the physical conditions of the gas in a variety of astrophysical environments (see further discussion below).194 Galaxies radiate a large percentage of (heir total luminosity in the infrared. from fens of percent in star formine galaxies up to in ultra luminous infrared. galaxies (Sandersetal.1938:Sanders&Mirabel1996).," Galaxies radiate a large percentage of their total luminosity in the infrared, from tens of percent in star forming galaxies up to in ultra luminous infrared galaxies \citep{Sanders88,Sanders96}."195.. An especially rich segment of the IR region falls between ~3 jan - 20 pom. where emission features from. PÀIIs arise in most galaxies.," An especially rich segment of the IR region falls between $\sim$ 3 $\mu$ m - 20 $\mu$ m, where emission features from PAHs arise in most star-forming galaxies."196 The emission from PAIIs varies from a lew percent up to of the total MIB. luminosity in normal galaxies (Ilelouetal.2000).. but (his percentage drops sharply with decreasing metallicitv (Engelbrachtetal.2005:Jackson 2007)..," The emission from PAHs varies from a few percent up to of the total MIR luminosity in normal galaxies \citep{Helou00}, but this percentage drops sharply with decreasing metallicity \citep{Engelbracht05,Jackson06,Smith07a}."197 Isolating the processes that affect PAIL emission intensity allows investigation of the physical conditions present in str forming regions (Ixennicuttetal.2003)., Isolating the processes that affect PAH emission intensity allows investigation of the physical conditions present in star forming regions \citep{Kennicutt03}.198. This in turn leads to enhanced understanding of the interplay between PAIL emission and star formation. the role of radiation field characteristics in governing PAIL intensities. and (he stellar-interstellar medium connection.," This in turn leads to enhanced understanding of the interplay between PAH emission and star formation, the role of radiation field characteristics in governing PAH intensities, and the stellar-interstellar medium connection."199 Alanw investigations have been conducted on PAIL emission in a wide range of galaxy types. but the observations of low-metallicitv galaxies in particular have been challenging. in lavee part due to the difficulties of obtaining spectra of sufficient sensitivity. ancl spatial resolution from low-mass (and often low surlace-brightness) galaxies.," Many investigations have been conducted on PAH emission in a wide range of galaxy types, but the observations of low-metallicity galaxies in particular have been challenging, in large part due to the difficulties of obtaining spectra of sufficient sensitivity and spatial resolution from low-mass (and often low surface-brightness) galaxies."200 Using Infrared. Array Camera (IRAC) imaging. Engelbrachtetal.(2005). discovered. a threshold metallicity of 12 + log(O/1I) zz 8. below whieh PAI intensity drops sharply. and these results have been confirmed by a number of other studies (e.g..Logeetal.2005:OHalloran tüherein)..," Using Infrared Array Camera (IRAC) imaging, \citet{Engelbracht05} discovered a threshold metallicity of 12 + log(O/H) $\approx$ 8, below which PAH intensity drops sharply, and these results have been confirmed by a number of other studies \citep[e.g.,][and references201 therein]{Hogg05,Ohalloran06,Jackson06,Rosenberg06,Engelbracht08,Wu10}."202 This metallicity apparently signifies a fundamental change in the nature of the interstellar medium (ISA); the molecular component of more metal-rich galaxies is easily, This metallicity apparently signifies a fundamental change in the nature of the interstellar medium (ISM); the molecular component of more metal-rich galaxies is easily203This work is partially supported by NSF of China 90103002 and the PhD Proeram Fuud of Chinese Education Ministry.,This work is partially supported by NSF of China 90103002 and the PhD Program Fund of Chinese Education Ministry.204 The authors wish to thank Dao-Neng Cao. Jiliaug Jiug. Jiau-Xiu Lu. and Shuane-Qine Wu for their stimulating discussious.," The authors wish to thank Dao-Neng Gao, Jiliang Jing, Jian-Xin Lu, and Shuang-Qing Wu for their stimulating discussions."205complex and Hermitian in à. Le. F5.m.)=FoU.m.6).,"complex and Hermitian in $\phi$, i.e. $F_{Q}^{*}(l,m,\phi)=F_{Q}(l,m,-\phi)$."206 We note further that Fy and Fy are not the local Stokes Q and U brightnesses at location (7.7.4). but simply auxiliary variables useful for the formalism.," We note further that $F_{Q}$ and $F_{U}$ are not the local Stokes $Q$ and $U$ brightnesses at location $(l,m,\phi)$, but simply auxiliary variables useful for the formalism."207 If we assume that F can be factored as in Eq. 12.. ," If we assume that $F$ can be factored as in Eq. \ref{eq:F_factorizable}, ,"208then We will now work only with Stokes Q. but we note that the following expressions also hold for Stokes Uf and Fy.," then We will now work only with Stokes $Q$, but we note that the following expressions also hold for Stokes $U$ and $F_U$."209 Following Eq. 3.. ," Following Eq. \ref{eq:true_to_measured_visibility}, ,"210the measured Stokes Q visibility. Vo. 1s where S is defined as in Eq. 4..," the measured Stokes $Q$ visibility, $\widehat{V_{Q}}$, is where $S$ is defined as in Eq. \ref{eq:uv_sampling_fn}."211 The relationship between the Stokes visibilities and the correlator output depends on the type of antenna feeds that are used., The relationship between the Stokes visibilities and the correlator output depends on the type of antenna feeds that are used.212 For linearly polarized feeds where Vyy. Vyy. ete.," For linearly polarized feeds where $V_{XX}$, $V_{YY}$, etc."213 represent the visibilities from cross-correlations of feeds having X or Y perpendicularly oriented dipoles., represent the visibilities from cross-correlations of feeds having X or Y perpendicularly oriented dipoles.214" For cireularly polarized antenna feeds. where Ve, and V; represent the visibilities from the cross-correlation between feeds of right and left. or left and right circular polarization. respectively."," For circularly polarized antenna feeds, where $V_{RL}$ and $V_{LR}$ represent the visibilities from the cross-correlation between feeds of right and left, or left and right circular polarization, respectively."215 We now define a(f.m.d) and (Lind) to be the representations of the primary beam and spectral dependence of the Faraday spectrum in Faraday space. respectively. 1e. and Using the convoletion theorem in between @ and ;U. we combine the expressions above to find that the Faraday spectrum is related to the measured visibilities by where ; denotes a 1D convolution with respect to ὁ along each LOS.," We now define $a(l,m,\phi)$ and $\sigma(l,m,\phi)$ to be the representations of the primary beam and spectral dependence of the Faraday spectrum in Faraday space, respectively, i.e. and Using the convolution theorem in between $\phi$ and $\lambda^{2}$, we combine the expressions above to find that the Faraday spectrum is related to the measured visibilities by where $\cnvphit$ denotes a 1D convolution with respect to $\phi$ along each LOS."216 This expression can be inverted to give the dirty image for the Faraday spectrum where and x is a full 3D convolution in /. 11. and Φ.," This expression can be inverted to give the dirty image for the Faraday spectrum where and $\cnvfullt$ is a full 3D convolution in $l$, $m$, and $\phi$."217 We note that the 3D and ID convolution operations do not commute., We note that the 3D and 1D convolution operations do not commute.218 We can see that by inverting the visibilities. we do not directly recover the Faraday spectrum. but rather the Faraday spectrum convolved with B in 3D. and c. and a in 1D along each LOS.," We can see that by inverting the visibilities, we do not directly recover the Faraday spectrum, but rather the Faraday spectrum convolved with $B$ in 3D, and $\sigma$, and $a$ in 1D along each LOS."219 In order to recover fo. one must first perform a 3D deconvolution using the CLEAN algorithm. for example.," In order to recover $f_{Q}$ , one must first perform a 3D deconvolution using the CLEAN algorithm, for example."220 After the 3D deconvolution. deconvolution of «à and « can be achieved by performing a ID inverse Fourier transform into 27-space along each LOS. dividing by A and s. and then Fourier transforming back into ó-space.," After the 3D deconvolution, deconvolution of $a$ and $\sigma$ can be achieved by performing a 1D inverse Fourier transform into $\lambda^2$ -space along each LOS, dividing by $A$ and $s$, and then Fourier transforming back into $\phi$ -space."221 The beam pattern A is usually known to high precision and is often represented by an analytic function parameterized in Lim. and ;U.," The beam pattern $A$ is usually known to high precision and is often represented by an analytic function parameterized in $l$, $m$, and $\lambda^{2}$."222 In addition. a map of s will be required.," In addition, a map of $s$ will be required."223 This can be obtained by measuring the spectral variation of total intensity maps along each LOS., This can be obtained by measuring the spectral variation of total intensity maps along each LOS.224 In many circumstances. it may be sufficient to simply assume that s is independent of / and i. since the errors introduced by using the wroο form for s are often quite small as previously discussed.," In many circumstances, it may be sufficient to simply assume that $s$ is independent of $l$ and $m$, since the errors introduced by using the wrong form for $s$ are often quite small as previously discussed."225 Imaging software that implements the 3D inversion given by Eq., Imaging software that implements the 3D inversion given by Eq.226 25 would avoid the complications described for traditional. 22-1D imaging.," \ref{eq:3D_inversion}227 would avoid the complications described for traditional, 2+1D imaging."228 We eliminate the eed to match incoverage at all frequencies because the 3D dirty beam. B. is constructed from the full 3D sampling function.," We eliminate the need to match $uv$ -coverage at all frequencies because the 3D dirty beam, $B$, is constructed from the full 3D sampling function."229 We also avoid the possibility of compounding errors through the process of deconvolving images that have already once been deconvolved., We also avoid the possibility of compounding errors through the process of deconvolving images that have already once been deconvolved.230 As a result. the fidelity of the images produced using Faraday synthesis should be improved over those made using the 2-1D technique.," As a result, the fidelity of the images produced using Faraday synthesis should be improved over those made using the 2+1D technique."231 In principle. the 3D approach will result η images that have higher dynamie range than those obtained using the 2+1D approach because we are able to use all data across the full bandwidth during imaging and deconvolution.," In principle, the 3D approach will result in images that have higher dynamic range than those obtained using the 2+1D approach because we are able to use all data across the full bandwidth during imaging and deconvolution."232 We have presented a simplified description of the Faraday spectrum measurement process above., We have presented a simplified description of the Faraday spectrum measurement process above.233 This will already work quite well in many circumstances. notably for narrow-fielc observations. without significant direction dependent effects. but in general additional steps will be required.," This will already work quite well in many circumstances, notably for narrow-field observations without significant direction dependent effects, but in general additional steps will be required."234 For instance. when the w-term in Eq.," For instance, when the $w$ -term in Eq."235 | eannot be ignored. the w-projectror algorithm of ? has been shown to be very effective in reducing imaging errors.," \ref{eq:vis_sky_relation_full} cannot be ignored, the $w$ -projection algorithm of \citet{cornwell_w_2005} has been shown to be very effective in reducing imaging errors."236 This algorithm makes use of the fact thatthe multiplication of the w-term and the sky brightness n1 the image plane is a convolution in visibility space., This algorithm makes use of the fact thatthe multiplication of the $w$ -term and the sky brightness in the image plane is a convolution in visibility space.237The w- visibilities are projected onto the wo=O plane,The $w$ -dependent visibilities are projected onto the $w=0$ plane238At the same time. the intermittency of the different MHD modes were shown to be very different.,"At the same time, the intermittency of the different MHD modes were shown to be very different."239 We clearly see the dependence of high order statistics of compressible motions on the Mach number., We clearly see the dependence of high order statistics of compressible motions on the Mach number.240 We interpret this dependence as the result of shock formation. which eventually changes the nature of the compressible motion cascade compared to the CLO3 assumptions.," We interpret this dependence as the result of shock formation, which eventually changes the nature of the compressible motion cascade compared to the CL03 assumptions."241 The limitations of the present study arise from the yet unclear nature of the turbulent cascade., The limitations of the present study arise from the yet unclear nature of the turbulent cascade.242 For instance. it was shown in that the degree of locality of interactions in hydrodynamic and MHD cascade are different.," For instance, it was shown in that the degree of locality of interactions in hydrodynamic and MHD cascade are different."243 Thus even largest available MHD simulations may not present the actual mertial range of the cascade. but the measured slope may be strongly affected by the extended bottle-neck effect of the simulations.," Thus even largest available MHD simulations may not present the actual inertial range of the cascade, but the measured slope may be strongly affected by the extended bottle-neck effect of the simulations."244 In addition. the limited range over which Alfvenic turbulence is weak may exhibit a rather different scaling of fast modes as a result of the interactions of the Alfvenic and fast modes 2005).," In addition, the limited range over which Alfvenic turbulence is weak may exhibit a rather different scaling of fast modes as a result of the interactions of the Alfvenic and fast modes ."245 In addition. within the present study we intentionally do not consider the scaling of magnetic perturbations.," In addition, within the present study we intentionally do not consider the scaling of magnetic perturbations."246 The velocity and magnetic perturbations for subAlfvenic turbulence show some differences. which are rather difficult to study reliably with the available numerical simulations.," The velocity and magnetic perturbations for subAlfvenic turbulence show some differences, which are rather difficult to study reliably with the available numerical simulations."247 These differences are not a part of the GS95 picture. but may reflect additional yet unclear properties of the MHD cascade 2000).," These differences are not a part of the GS95 picture, but may reflect additional yet unclear properties of the MHD cascade ."248. In our study we used only the incompressible driving., In our study we used only the incompressible driving.249 In the presence of the compressible supersonic driving the scaling looks different. but the existence of the inertial range is then questionable.," In the presence of the compressible supersonic driving the scaling looks different, but the existence of the inertial range is then questionable."250 claimed that combining the compressible and incompressible driving in the Mach number dependent fashion one can obtain a better power-law inertial range., claimed that combining the compressible and incompressible driving in the Mach number dependent fashion one can obtain a better power-law inertial range.251 This issue requires further studies., This issue requires further studies.252 The turbulence driving in our study is balanced. in the sense that the energy flows in opposite directions are equal.," The turbulence driving in our study is balanced, in the sense that the energy flows in opposite directions are equal."253 In the presence of sources and sinks of turbulent energy. astrophysical turbulence is expected to be imbalanced.," In the presence of sources and sinks of turbulent energy, astrophysical turbulence is expected to be imbalanced."254 Our numerical studies of imbalanced turbulence in. show that the properties of Alfvenic turbulence changes substantially in the presence of imbalance., Our numerical studies of imbalanced turbulence in show that the properties of Alfvenic turbulence changes substantially in the presence of imbalance.255 However. the degree of sustainable imbalance in compressible turbulence is still unclear.," However, the degree of sustainable imbalance in compressible turbulence is still unclear."256 One expects the density fluctuation in turbulent fluid to reflect the incoming waves. altering the imbalance.," One expects the density fluctuation in turbulent fluid to reflect the incoming waves, altering the imbalance."257 We believe that in high Mach number fluids the imbalance is low due to the existence of substantial density contrasts., We believe that in high Mach number fluids the imbalance is low due to the existence of substantial density contrasts.258 Depending on driving astrophysical turbulence may be subAlfvénnic. if the injection velocity V; is less than Alfven speed V4. Alfvenic. if Vj;=V4. and superAlfvenic. if Vj>V4.," Depending on driving astrophysical turbulence may be subAlfvénnic, if the injection velocity $V_L$ is less than Alfven speed $V_A$, Alfvenic, if $V_L=V_A$, and superAlfvenic, if $V_L>V_A$."259" This frequently is also deseribed by the Alfven Mach number M,=Vi/V4.", This frequently is also described by the Alfven Mach number $M_A=V_L/V_A$.260 Formally. the GS95 model applies only to incompressible motions with Vy=V4. or equivalently M4=1.," Formally, the GS95 model applies only to incompressible motions with $V_L=V_A$, or equivalently $M_A=1$."261 Some of the astrophysical applications of the model. indeed. use the original form of the theory. which substantially limits the applications of the theory2001).," Some of the astrophysical applications of the model, indeed, use the original form of the theory, which substantially limits the applications of the theory."262 However. the model can be easily generalized to cover extensive ranges of superAlfvenic and subAlfvenic turbulence2006).," However, the model can be easily generalized to cover extensive ranges of superAlfvenic and subAlfvenic turbulence."263" For subAlfvenic turbulence with isotropic driving at the scale L an initial weak cascade. in. which the parallel scale of motions stays the same and the spectrum E(K,)~KT Is applicable. transfers to the regime of strong turbulence at the scale of LM;. for which the GS95 eritical balance arguments are applicable."," For subAlfvenic turbulence with isotropic driving at the scale $L$ an initial weak cascade, in which the parallel scale of motions stays the same and the spectrum $E(k_{\bot})\sim k_{\bot}^{-2}$ is applicable, transfers to the regime of strong turbulence at the scale of $LM_A^2$, for which the GS95 critical balance arguments are applicable."264 For superAlfvente turbulence. while up to the scale LM? the turbulence is hydrodynamic. it approaches the GS95-type regime for smaller scales.," For superAlfvenic turbulence, while up to the scale $LM_A^{-3}$ the turbulence is hydrodynamic, it approaches the GS95-type regime for smaller scales."265 Therefore. the relations obtained for MHD turbulence that we have studied above can be generalized for cases of different intensity of driving.," Therefore, the relations obtained for MHD turbulence that we have studied above can be generalized for cases of different intensity of driving."266 While the GS95 model is a model of incompressible turbulence. our simulations confirm the numerical findings in and CLO3 that the scaling of the Alfvenic mode in the compressible turbulence is very similar to its scaling in the incompressible case.," While the GS95 model is a model of incompressible turbulence, our simulations confirm the numerical findings in and CL03 that the scaling of the Alfvenic mode in the compressible turbulence is very similar to its scaling in the incompressible case."267 In particular. the GS95 anisotropy of MHD turbulence determines the rate of magnetic field wandering which 15 important for many astrophysical processes. including the ubiquitous process of magnetic reconnection1999).," In particular, the GS95 anisotropy of MHD turbulence determines the rate of magnetic field wandering which is important for many astrophysical processes, including the ubiquitous process of magnetic reconnection."268. Additional implications of magnetic field wandering include the diffusion of heat and cosmic rays. MHD acceleration of dust ete.," Additional implications of magnetic field wandering include the diffusion of heat and cosmic rays, MHD acceleration of dust etc."269review)... The wavelet approach has the potential of increasing accuracy while studying small-scale anisotropy in. simulations with strongly perturbed magnetic fields., The wavelet approach has the potential of increasing accuracy while studying small-scale anisotropy in simulations with strongly perturbed magnetic fields.270 and collaborators attracted. the attention of the interstellar community to the potential important implications of intermittency., and collaborators attracted the attention of the interstellar community to the potential important implications of intermittency.271 A small and transient volume with high temperatures or violent turbulence can have significant effects on the net rates of processes within the ISM., A small and transient volume with high temperatures or violent turbulence can have significant effects on the net rates of processes within the ISM.272 For instance. many interstellar chemical reactions (e.g. the strongly endothermic formation of CH*) might take place within very intensive intermittent vortices.," For instance, many interstellar chemical reactions (e.g., the strongly endothermic formation of $^+$ ) might take place within very intensive intermittent vortices."273 The aforementioned authors claimed the existence of the observational evidence for such reactions and heating. but a more quantitative approach to the problem is possible.," The aforementioned authors claimed the existence of the observational evidence for such reactions and heating, but a more quantitative approach to the problem is possible."274 used the intermittency scaling and calculated the distribution of the dissipation rate in. the turbulent volumes., used the intermittency scaling and calculated the distribution of the dissipation rate in the turbulent volumes.275 In doing so they used the fact that model of intermittency corresponds generalized log-Poisson distribution of the local dissipation rates(1995)., In doing so they used the fact that model of intermittency corresponds generalized log-Poisson distribution of the local dissipation rates.276. The obtained rates of enhancement were not sufficient to explain the heating required for inducing interstellar chemistry2007)., The obtained rates of enhancement were not sufficient to explain the heating required for inducing interstellar chemistry.277. The same approach was used by who obtained. however. a different result.," The same approach was used by who obtained, however, a different result."278 We believe that one should distinguish shocks from. vortical motions while calculating the heating induced by intermittency., We believe that one should distinguish shocks from vortical motions while calculating the heating induced by intermittency.279 Our present study show very different scalings relevant to these types of motions., Our present study show very different scalings relevant to these types of motions.280 Numerous studies of compressible MHD turbulence are done in the context of star formationtherein)., Numerous studies of compressible MHD turbulence are done in the context of star formation.281. Most of these simulations are focused on the large- appearances of the turbulence. which is determined by the turbulent driving and do not exhibit any extended inertial range of turbulence.," Most of these simulations are focused on the large-scale appearances of the turbulence, which is determined by the turbulent driving and do not exhibit any extended inertial range of turbulence."282 Search for the universal relations for compressible turbulence resulted in the rise of interest to the idea of searching universality not for velocity. but for the combination of the velocity and density in the form," Search for the universal relations for compressible turbulence resulted in the rise of interest to the idea of searching universality not for velocity, but for the combination of the velocity and density in the form"28355 reveals (hat (heprojected radial density profiles of NC merger remnants. Xxc(AR). are sienificantly different between our four models with different. pj.,"5 reveals that the radial density profiles of NC merger remnants, ${\Sigma}_{\rm NC}(R)$, are significantly different between our four models with different $F_{\rm BH}$."284 The rather low central Vve value al R=0.0560 and shallow inner density profile. in the model with £j=0.05 suggests that if this merger remnant is located in the central region of a galaxy. it is less likely to be observed as a distinct NC.," The rather low central ${\Sigma}_{\rm NC}$ value at $R=0.05R_{\rm NC}$ and shallow inner density profile, in the model with $F_{\rm BH}=0.05$ suggests that if this merger remnant is located in the central region of a galaxy, it is less likely to be observed as a distinct NC."285 While the order of magnitude drop in surface density nav seen like overkill. especially given (he apparently small levels of excess nuclear light seen in most resolution-limited images. we note that well-resolved galaxies can have NC light up to 5 mag 7? (LOOX) brighter than the underlving galaxy (e.g... Graham Spitler 2009).," While the order of magnitude drop in surface density may seem like overkill, especially given the apparently small levels of excess nuclear light seen in most resolution-limited images, we note that well-resolved galaxies can have NC light up to 5 mag $^{-2}$ $\times$ ) brighter than the underlying galaxy (e.g., Graham Spitler 2009)."286 The present study. confirms that more evolved NC's by which we mean NC's. with MBUs. that are farther along the merger (vee can have lower inner densities (pxc and Xx) and shallower inner 2D density profiles (han their progenitors.," The present study confirms that more evolved NCs — by which we mean NCs, with MBHs, that are further along the merger tree — can have lower inner densities ${\rho}_{\rm NC}$ and ${\Sigma}_{\rm NC}$ ) and shallower inner 2D density profiles than their progenitors."287 This suggests (hat boundaries between clistinet stellar nuclei and background5 field stars in 5galaxies are less clear [or more evolved svstems as the NCs are effectively. washed-out ancl dissolve into the host galaxy., This suggests that boundaries between distinct stellar nuclei and background field stars in galaxies are less clear for more evolved systems as the NCs are effectively washed-out and dissolve into the host galaxy.288 such diffuse NCs are also more susceptible to tidal destruction during galaxy merging., Such diffuse NCs are also more susceptible to tidal destruction during galaxy merging.289 The present study. therefore suggests that the observed. /5jj- Mi relation can be understood in terms of the structural evolution of merging NCs with AIBIIs.," The present study therefore suggests that the observed $f_{\rm290 BH}$ $M_{\rm sph}$ relation can be understood in terms of the structural evolution of merging NCs with MBHs."291 Measurements of partiallv-depleted galaxy cores. relative to a galaxys outer light-prolile. have revealed a correlation between the central stellar mass deficit and the luminosity of the host spheroid and its MBIT mass (e.g.. Graham 2004: Ferrarese et 22006b).," Measurements of partially-depleted galaxy cores, relative to a galaxy's outer light-profile, have revealed a correlation between the central stellar mass deficit and the luminosity of the host spheroid and its MBH mass (e.g., Graham 2004; Ferrarese et 2006b)."292" As detailed in Graham Guzmann (2003) and Cotté et ((2007). the transition between massive ealaxies with parlially-depletecl cores and those without which frequently have excess nuclear light. instead occurs around Af,=—20.5 mag."," As detailed in Graham Guzmánn (2003) and Côtté et (2007), the transition between massive galaxies with partially-depleted cores and those without — which frequently have excess nuclear light instead — occurs around $M_B =293-20.5$ mag."294 Previous numerical simulations proposed that the origm of these central stellar deficits can be understood in the context of core formation through dwnamical heating of stars by inspiralling MDlIIs in galaxy merging (e.g.. Ebisuzaki et al.," Previous numerical simulations proposed that the origin of these central stellar deficits can be understood in the context of core formation through dynamical heating of stars by inspiralling MBHs in galaxy merging (e.g., Ebisuzaki et al."295 1991)., 1991).296 The present study has. for the first tme. addressed one of the over-looked problems related to the nuclear structures of galaxies: why. and how can dense NC's disappear during galaxy growth through galaxy merging?," The present study has, for the first time, addressed one of the over-looked problems related to the nuclear structures of galaxies: why and how can dense NCs disappear during galaxy growth through galaxy merging?"297 We advocate here that core-depletion due to the gravitational slingshot of host galaxy stars by inspiralling MDIIs will not occur in earnest until the NC's surrounding the MDIIs have first been eroded away by (his same mechanism: once the NC's are effectively gone. the binary MBIIs. perhaps from additional merger events. can then commence to eat into the inner lieht profile of the host galaxy. to produce the observed. parlially-depleted cores.," We advocate here that core-depletion due to the gravitational slingshot of host galaxy stars by inspiralling MBHs will not occur in earnest until the NCs surrounding the MBHs have first been eroded away by this same mechanism: once the NCs are effectively gone, the binary MBHs, perhaps from additional merger events, can then commence to eat into the inner light profile of the host galaxy to produce the observed partially-depleted cores."298" This important step can explain why NC's disappear along the spheroid mass sequence and il also olfers a process through which to understand the observed £14,744 relationship in terms ol galaxy formation within the hierarchical merging scenario."," This important step can explain why NCs disappear along the spheroid mass sequence and it also offers a process through which to understand the observed $F_{\rm BH}$ $M_{\rm299bulge}$ relationship in terms of galaxy formation within the hierarchical merging scenario."300In conclusion. the //ST imaging 87. 304 and 680d aller V-band maximum sav little about circumstellar material closer than about 1.3κ107 cm.,"In conclusion, the $HST$ imaging 87, 304 and 680d after V-band maximum say little about circumstellar material closer than about $1.3\times 10^{18}$ cm."301 As in the case of SN 1991T and 1995E. no central source corresponding to a circumstellar structure. echoing or not. has ever been detected.," As in the case of SN 1991T and 1995E, no central source corresponding to a circumstellar structure, echoing or not, has ever been detected."302 Ii contrast. SN 1993bu shows a central source that has been interpreted as a circumstellar echo.," In contrast, SN 1998bu shows a central source that has been interpreted as a circumstellar echo."303 We note. however. that an image-subiraction comparison of the existing archival //ST imaging shows verv little change in this source between vears 2000 and 2006.," We note, however, that an image-subtraction comparison of the existing archival $HST$ imaging shows very little change in this source between years 2000 and 2006."304 We are investigating if spectroscopy of this source will reveal it as an echo. a supernova renimant. or some other structure (Crotis 2008).," We are investigating if spectroscopy of this source will reveal it as an echo, a supernova remnant, or some other structure (Crotts 2008)."305 To summarize. there are a number of significant results that these new observations have produce: 1) The echo signal arises primarily [rom a prestunably interstellar sheet. of. material 26.32:3.2 pe in the SN foreground. and this structure likely represents the dominant portion ol the large amount of extinction along the SN sightline.," To summarize, there are a number of significant results that these new observations have produced: 1) The echo signal arises primarily from a presumably interstellar sheet of material $\pm$ 3.2 pc in the SN foreground, and this structure likely represents the dominant portion of the large amount of extinction along the SN sightline."306 2) As of 680 days after V-band maximum light there is no indication of anv cireumstellar echo (at anv level above of the total echo signal). and the imaging and photometry at 304 days is consistent with these results. both in the presence of an interstellar echo of similar geometry and brightness. and the absence of evidence of a cireiumstellar echo.," 2) As of 680 days after V-band maximum light there is no indication of any circumstellar echo (at any level above of the total echo signal), and the imaging and photometry at 304 days is consistent with these results, both in the presence of an interstellar echo of similar geometry and brightness, and the absence of evidence of a circumstellar echo."307 3) There is no evidence of any circumstellar echo signal al any epoch in these data. but thev do not bear directly on the plausibly cireumstellar absorption signal seen in Na I making SN 2006X a candidate for one of the few Type Ia supernovae in which cireumstellar matter has been knowingly detected.," 3) There is no evidence of any circumstellar echo signal at any epoch in these data, but they do not bear directly on the plausibly circumstellar absorption signal seen in Na I making SN 2006X a candidate for one of the few Type Ia supernovae in which circumstellar matter has been knowingly detected."308 [thank Steve Lawrence and Den Sugerman for helpful discussion., I thank Steve Lawrence and Ben Sugerman for helpful discussion.309 This work was supported bv erants GO/DD-10991 and GO 11171 from οΕΟΤ. Aldering. G.. et 22006. ApJ. 650. 510.," This work was supported by grants GO/DD-10991 and GO 11171 from STScI. Aldering, G., et 2006, ApJ, 650, 510."310 Altavilla. G.. et 22004. AINRAS. 349. 1344.," Altavilla, G., et 2004, MNRAS, 349, 1344."311 Benetli. ο. et 22006. ApJ. 653. L129.," Benetti, S., et 2006, ApJ, 653, L129."312 Chandra. P... Chavalier.," Chandra, P., Chavalier,"313"be where 75544, is the lensing signal. ¢intrinsie 15 the noise due to the intrinsic shape of the galaxies, ego; 18 the statistical error in the measurement from the photon shot noise, and epsy is the error in the shear estimate due to uncorrected PSF contamination.","be where $\gamma_{\rm grav}$ is the lensing signal, $\epsilon_{\rm314intrinsic}$ is the noise due to the intrinsic shape of the galaxies, $\epsilon_{\rm meas}$ is the statistical error in the measurement from the photon shot noise, and $\epsilon_{\rm PSF}$ is the error in the shear estimate due to uncorrected PSF contamination."315" The systematic errors [rom the PSF interpolation enter through the term epa, which arises from errors in the PSF ellipücity."," The systematic errors from the PSF interpolation enter through the term $\epsilon_{\rm PSF}$, which arises from errors in the PSF ellipticity."316" Using equation 1.. which expands it in principal components, we can wrile eps as: where à refers to the error in the estimate of a quantity, and the last sum includes all of the patterns which are not modeled by the PCA, including any completely random effects which do not recur in multiple exposures."," Using equation \ref{eqn:psf}, which expands it in principal components, we can write $\epsilon_{\rm PSF}$ as: where $\delta$ refers to the error in the estimate of a quantity, and the last sum includes all of the patterns which are not modeled by the PCA, including any completely random effects which do not recur in multiple exposures."317" A represents the conversion from ellipicity to shear, which Bernstein&Jarvis(2002). refer to asresponsivily!."," ${\cal R}$ represents the conversion from ellipticity to shear, which \citet{BJ02} refer to as."318. When using this technique for other properties of the PSF besides ellipticity (size for example). A. would be the corresponding mean effect that errors in the measurement have on the net shear estimates Irom the galaxy shapes.," When using this technique for other properties of the PSF besides ellipticity (size for example), ${\cal R}$ would be the corresponding mean effect that errors in the measurement have on the net shear estimates from the galaxy shapes."319" We will refer to the estimates of the two-point correlation funcuon [rom observations of galaxies on (wo exposures, 7 and j as: where we omit the + and — subscripts, both here and in much of the further discussion, leaving the appropriate conjugation or not in the two cases implied."," We will refer to the estimates of the two-point correlation function from observations of galaxies on two exposures, $i$ and $j$ as: where we omit the $+$ and $-$ subscripts, both here and in much of the further discussion, leaving the appropriate conjugation or not in the two cases implied."320 The statistical errors Irom the measurement noise and intrinsic ellipiiciües are well understood., The statistical errors from the measurement noise and intrinsic ellipticities are well understood.321" We now look at what can contribute to the systematic PSF contamination, eps, and how that propagates to ελλ,"," We now look at what can contribute to the systematic PSF contamination, $\epsilon_{\rm PSF}$, and how that propagates to $\hat\xi^{(i,j)}$."322" The errors in the three-point function are completely analogous, so it is sufficient to only refer to the two-point function here."," The errors in the three-point function are completely analogous, so it is sufficient to only refer to the two-point function here."323Ol the 81FUSE locations. 43 overlapped with 10CYT field of observations in the LMC and we have found a strong correlation between.FUSE and.CIT diffuse flix (Fig. 3)).,"Of the 81 locations, 43 overlapped with 10 field of observations in the LMC and we have found a strong correlation between and diffuse flux (Fig. \ref{Fig3}) ),"324 where the UIT fluxes have been integrated. over the x FUSE aperture., where the UIT fluxes have been integrated over the x FUSE aperture.325 Three points in (he 30 Doradus region have a relatively higher FUSE flix because the radiation field is dominated by O stars in the nebula whereas one observation in Nil had a relatively higher UIT flux because the stellar spectrum is much flatter (han in other regions., Three points in the 30 Doradus region have a relatively higher FUSE flux because the radiation field is dominated by O stars in the nebula whereas one observation in N11 had a relatively higher UIT flux because the stellar spectrum is much flatter than in other regions.326 Rather than deal with absolute values. a useful comparison is to find (he Iractional amount of diffuse radiation in the field defined as diffuse radiation over total radiation (diffuse + stellar) in each of the UIT regions. as was done by Parkeretal.(1993).," Rather than deal with absolute values, a useful comparison is to find the fractional amount of diffuse radiation in the field defined as diffuse radiation over total radiation (diffuse + stellar) in each of the UIT regions, as was done by \citet{Parker98}."327.. We used the stars [rom his cabalog. (translated (hem into the FUSE spectral range assuming lvurucz models and summed the fIuxes in a given field.," We used the stars from his catalog, translated them into the FUSE spectral range assuming Kurucz models \citep{Kurucz92} and summed the fluxes in a given field."328 With the observed FUSE/UIT diffuse ratio. we could then caleulate the total amount of diffuse radiation in each of the FUSE bands and (hus the fraction of total light emitted as diffuse emission.," With the observed FUSE/UIT diffuse ratio, we could then calculate the total amount of diffuse radiation in each of the FUSE bands and thus the fraction of total light emitted as diffuse emission."329 These fractions range Iron to of the total at 1100A.. with a high of in the superbubble NTO. with an observed error of 12. 17.," These fractions range from to of the total at 1100, with a high of in the superbubble N70, with an observed error of 12 –."330. The comparable estimate for the Milkv. Wavy is ).., The comparable estimate for the Milky Way is \citep{Parravano03}.331 Although some part of the variation of diffuse fraction in different regions of the LMC may be due to the dust distribution. it is likely that much of the stellar radiation is as noted bv Coleetal.(1999a):: r.e.. the diffuse light may come from stars far away from the observed area.," Although some part of the variation of diffuse fraction in different regions of the LMC may be due to the dust distribution, it is likely that much of the stellar radiation is non-local, as noted by \citet{Cole99a}; i.e., the diffuse light may come from stars far away from the observed area."332 Ir our Galaxy. (his is seen as scattering of galactic plane star light bv high latitude cust clouds (Jura1980): in the LMC. light from the OB associations will be scattered by distant dust.," In our Galaxy, this is seen as scattering of galactic plane star light by high latitude dust clouds \citep{Jura80}; in the LMC, light from the OB associations will be scattered by distant dust."333 The shape of the diffuse fraction (diffuse radiation over total radiation) is essentially the same in all the 10 regions (Fig. 4)), The shape of the diffuse fraction (diffuse radiation over total radiation) is essentially the same in all the 10 regions (Fig. \ref{Fig4}) )334 rising by a [actor of about 5 from 1000 to 1500À.. implving that most of the heating of the interstellar dust comes in the FUV.," rising by a factor of about 5 from 1000 to 1500, implying that most of the heating of the interstellar dust comes in the FUV."335 This is consistent with the lower albedo (dashed line) and higher cross-section (dot-dash line) of the grains in the FUV., This is consistent with the lower albedo (dashed line) and higher cross-section (dot-dash line) of the grains in the FUV.336 Coleetal.(1999b)} have attempted to model the distribution of diffuse light in (he1) data by scattering the light of OD associations in the LMC from a dust distribution which decays exponentially with radius aud wilh a hyvperbolie secant with distance from the plane., \citet{Cole99b} have attempted to model the distribution of diffuse light in the data by scattering the light of OB associations in the LMC from a dust distribution which decays exponentially with radius and with a hyperbolic secant with distance from the plane.337 Thev found that although (heir models did match the overall morphology of the observations. ib was difficult to constrain the parameters because of the uncertainty in many of the physical properties of the ISM in the LMC. particularly in its clumping.," They found that although their models did match the overall morphology of the observations, it was difficult to constrain the parameters because of the uncertainty in many of the physical properties of the ISM in the LMC, particularly in its clumping."338 Nevertheless they did find that 30 Doracdus dominated the diffuse emission in the eastern LMC. a conclusion borne out by ow FUSE observations in ihe FUV.," Nevertheless they did find that 30 Doradus dominated the diffuse emission in the eastern LMC, a conclusion borne out by our observations in the FUV."339"ones will not be detectable. however. with current sensitivities,","ones will not be detectable, however, with current sensitivities."340 On the other haud. those with lower iutrinsic redshifts. which have been found to be more huumous than those with high intrinsic redshifts (Bell2002a.c).. will be detectable to ereater cosmological distances.," On the other hand, those with lower intrinsic redshifts, which have been found to be more luminous than those with high intrinsic redshifts \citep{bel02a,bel02c}, will be detectable to greater cosmological distances."341 The cutive sample coutaimine 11.200 redshifts is plotted in Fig 2.," The entire sample containing 44,200 redshifts is plotted in Fig 2."342 As in Fie Ll. from equ 2. the vertical dashed lines indicate the highest redshift level in each N-eroup aud the dotted lines indicate the lower levels inside cach eroup.," As in Fig 1, from eqn 2, the vertical dashed lines indicate the highest redshift level in each $N$ -group and the dotted lines indicate the lower levels inside each group."343 Where the region above z = 2.3 contaius ouly two lines iu cach eroup. as can be seen in Fie.," Where the region above z = 2.3 contains only two lines in each group, as can be seen in Fig."344 2 and iu Table 1. below z = 2 the density of predicted intrinsic redshift lines inside cach group increases sienificautly.," 2 and in Table 1, below z = 2 the density of predicted intrinsic redshift lines inside each group increases significantly."345 Tuunuediatelv evident in Fig 2 is the fact that he density of SDSS redshifts iu the lower half of he sample is wich higher than in the upper halt. although part of this difference can be attributed o the low-vedshift sample with (mae cut-off at / = 19.1.," Immediately evident in Fig 2 is the fact that the density of SDSS redshifts in the lower half of the sample is much higher than in the upper half, although part of this difference can be attributed to the low-redshift sample with $i$ -mag cut-off at $i$ = 19.1."346 Similarly. there is a 1iuch higher density of oxedieted lines in the lower region. especially for he model described by equ 2.," Similarly, there is a much higher density of predicted lines in the lower region, especially for the model described by eqn 2."347 Although. for equ 2. the higher line density below z = 2 introduces sone confusion. there is still evidence for peaks hat correspoud tothe first three /N-eroups located at. or near. z = 0.62. L.21. aud. 1.86. depenudiug ou the liue distribution inside each group.," Although, for eqn 2, the higher line density below z = 2 introduces some confusion, there is still evidence for peaks that correspond to the first three $N$ -groups located at, or near, z = 0.62, 1.24, and 1.86, depending on the line distribution inside each group."348 Ouly at z = 2.18. where the source deusity is rising steeply. is there no obvious peak. although the DR1 distribution in Fig 1 does show onc.," Only at z = 2.48, where the source density is rising steeply, is there no obvious peak, although the DR1 distribution in Fig 1 does show one."349 The lack of a clear peak at zg = 2.18 is not surprising since the survey portion with /auag cut-off at / = 19.1 found mainly sources below z ~+2.2., The lack of a clear peak at z = 2.48 is not surprising since the survey portion with $i$ -mag cut-off at $i$ = 19.1 found mainly sources below z $\sim$ 2.2.350 This has contributed to the sharp drop in source density above z = 2.2. aud such an abrupt cliauge in level can easilv mask a weak peak.," This has contributed to the sharp drop in source density above z = 2.2, and such an abrupt change in level can easily mask a weak peak."351 Iu Fig 3 the correlation coefficient r. between the actual redshift distribution aud the distribution of redshifts predicted using equ 2 is plotted versus shift.," In Fig 3 the correlation coefficient r, between the actual redshift distribution and the distribution of redshifts predicted using eqn 2 is plotted versus shift."352 It shows a correlation cocficient of 0.12 for zero shift., It shows a correlation coefficient of 0.42 for zero shift.353 For a distribution coutaimine [5 salples as here. r greater than 0.288 is considered sieuificaut.," For a distribution containing 45 samples as here, r greater than 0.288 is considered significant."354 IHTowever. more iiportautlv. as the two distributions are shifted relative to cach other. r first decreases aud then increases again at x6 biu," However, more importantly, as the two distributions are shifted relative to each other, r first decreases and then increases again at $\pm6$ bin"355field). there is also a significant range of regular. orbits.,"field), there is also a significant range of regular orbits."356 Aloreover. since in our case we deal with finite disc models. a distinction. between disc-crossing ancl non-clise-crossing orbits is sometimes necessary.," Moreover, since in our case we deal with finite disc models, a distinction between disc-crossing and non-disc-crossing orbits is sometimes necessary."357 In the latter case we did not find chaotic motion. even in extreme situations where there are saddle points outside the field source. and the disc and cisc-[ree region are connected (Figures 13. ancl 14)).," In the latter case we did not find chaotic motion, even in extreme situations where there are saddle points outside the field source, and the disc and disc-free region are connected (Figures \ref{Akalnajs1} and \ref{Akalnajs2}) )."358 Although one would be tented to think that such hyperbolic exterior. points can induce chaos at disc-free regions. what really happens is that the stochastic motion tends toward dise regions and a completely regular motion is developed outside there.," Although one would be tented to think that such hyperbolic exterior points can induce chaos at disc-free regions, what really happens is that the stochastic motion tends toward disc regions and a completely regular motion is developed outside there."359 “Phese considerations have special relevance in galaxy models with thin cise plus halo components., These considerations have special relevance in galaxy models with thin disc plus halo components.360 Particles belonging to the halo component will follow a motion with the features mentioned above and. as it was showed by some authors. this fact determines decisively the internal structure of such stellar systems (see Ostriker. Spitzer and Chevalier (1972))).," Particles belonging to the halo component will follow a motion with the features mentioned above and, as it was showed by some authors, this fact determines decisively the internal structure of such stellar systems (see Ostriker, Spitzer and Chevalier \citeyear{Ostriker}) ))."361 The authors thank to Leonardo Pachónn for his valuable sugeestions ancl orientations., The authors thank to Leonardo Pachónn for his valuable suggestions and orientations.362 Ci. A. CG. and E. L-S. want to thank the financial support. from COLCTIISNCITAS. Colombia. whereas that J. R-C. and EF. L-S want to thank the finantial support fromAcadémica. Universidad. Industrial de Santander.," G. A. G. and F. L-S. want to thank the financial support from COLCIENCIAS, Colombia, whereas that J. R-C. and F. L-S want to thank the finantial support from, Universidad Industrial de Santander."363where σερ). is defined by equating As) to su.,where $z_{spi}(z)$ is defined by equating $\Delta(z)$ to $\Delta_{spi}$.364 Finally. SO galaxies are such that: Figure 1 illustrates the way morphological types are defined in this model.," Finally, S0 galaxies are such that: Figure \ref{figcutsz} illustrates the way morphological types are defined in this model."365 Phe formation redshift: is shown as a function of the epoch + considered., The formation redshift is shown as a function of the epoch $z$ considered.366 “Phe solid. curves are obtained for (2)=Avy (upper) and Af.)=au; (lower)., The solid curves are obtained for $\Delta(z)=\Delta_{ell}$ (upper) and $\Delta(z)=\Delta_{spi}$ (lower).367" Ellipticals occupy the upper region. of the (2,5:.2) plane. while disks occupy the lower region."," Ellipticals occupy the upper region of the $z_{nl},z$ ) plane, while disks occupy the lower region."368 SOs are located. in between., S0s are located in between.369 The hatched area corresponds to tn;«2 and is irrelevant., The hatched area corresponds to $z_{nl}<z$ and is irrelevant.370 Consider the example of a galaxy that became. linear at. sav. 24;=3.," Consider the example of a galaxy that became non-linear at, say, $z_{nl}=3$."371 Figure 1. shows that up to redshift 2o208 this galaxy is identified as a disk., Figure \ref{figcutsz} shows that up to redshift $z\sim 2.8$ this galaxy is identified as a disk.372 Between 2~2.2 and z~2.8. the same galaxy has experienced: sullicient energy exchange with its neighbours to be identified as an SO galaxy. but not quite enough to be an elliptical.," Between $z\sim 2.2$ and $z\sim 2.8$, the same galaxy has experienced sufficient energy exchange with its neighbours to be identified as an S0 galaxy, but not quite enough to be an elliptical."373 By 2~2.2. it has become an elliptical.," By $z\sim 2.2$, it has become an elliptical."374 Depending on the previously identified: morphological type. we decide which galaxies will undergo an “obscure starburst”.," Depending on the previously identified morphological type, we decide which galaxies will undergo an “obscure starburst”."375 In other words. this phase is triggered when the enerey imparted by encounters/collisions is strong enough to alter the morphology for the first time. Le. when late-tvpe (spiral) 3galaxies first become carly tvpe (SOs).," In other words, this phase is triggered when the energy imparted by encounters/collisions is strong enough to alter the morphology for the first time, i.e. when late-type (spiral) galaxies first become early type (S0s)."376 Phe transition from SO to elliptical is then supposed to happen smoothly. without another burst of star formation. as the gas poor SO galaxies are assumed. to have reheated their surroundings enough during the burst to not be able to replenish their gas supplies through accretion of new material.," The transition from SO to elliptical is then supposed to happen smoothly, without another burst of star formation, as the gas poor SO galaxies are assumed to have reheated their surroundings enough during the burst to not be able to replenish their gas supplies through accretion of new material."377 The intensity ancl duration of this LIRG/ULIRG phases. are therefore controlled by the amount of eas available for star formation and the size of the galaxy at the time when the morphology change occurs.," The intensity and duration of this LIRG/ULIRG phases, are therefore controlled by the amount of gas available for star formation and the size of the galaxy at the time when the morphology change occurs."378" In. practice. this dark phase is nioclelled by setting our three Κον parameters to 7=1. f,=1 and (gv=0.5. corresponding to high star formation elficiency. high dust opacity. and high feedback elliciency. respectively."," In practice, this dark phase is modelled by setting our three key parameters to $\beta = 1$, $f_c = 1$ and $\epsilon_{SN} = 0.5$, corresponding to high star formation efficiency, high dust opacity, and high feedback efficiency respectively."379 Lt is therefore completely. coupled to ie starburst. and as the lumunosity of this latter decreases for lack οἱ fuel (eas) supply. the optical depth of the galaxy goes down as well. revealing more and more of the stellar population at optical wavelongths.," It is therefore completely coupled to the starburst, and as the luminosity of this latter decreases for lack of fuel (gas) supply, the optical depth of the galaxy goes down as well, revealing more and more of the stellar population at optical wavelengths."380 Looking at figures 2. and 3.. one realises that (except. in the near LR bands). late type galaxies dominate over carly types (compare the dotted curves with the dashed ones in each. panel).," Looking at figures \ref{figopt} and \ref{figinf}, one realises that (except in the near IR bands), late type galaxies dominate over early types (compare the dotted curves with the dashed ones in each panel)."381 This domination extends down to the fu-Lt. with the late-tvpe galaxies still dominating the 175 micron ISOPLIOT counts.," This domination extends down to the far-IR, with the late-type galaxies still dominating the 175 micron ISOPHOT counts."382 At longer wavelengths. however. there is à dramatic change: the earlv-tvpe. galaxies completely swamp the contribution from late types.," At longer wavelengths, however, there is a dramatic change: the early-type galaxies completely swamp the contribution from late types."383 Indeed. one can see on the bottom right panel of figure 3. that the vast majority (2 90 '4)) of the SCUBA sources are. classified as carly tvpes in our model.," Indeed, one can see on the bottom right panel of figure \ref{figinf} that the vast majority $\approx$ 90 ) of the SCUBA sources are classified as early types in our model."384 Phe reason for such a change of behaviour lies in the well known negative k-correction. which makes galaxies of the same bolometric luminosity as bright for the observer at. redshift 5 as at reclshilt 0.5.," The reason for such a change of behaviour lies in the well known negative k-correction, which makes galaxies of the same bolometric luminosity as bright for the observer at redshift 5 as at redshift 0.5."385 This is only important in the submm (here for SCUBA at S50 microns). because the peak cniissivity of dust in the source rest frame is between 60 ancl LOO microns.," This is only important in the submm (here for SCUBA at 850 microns), because the peak emissivity of dust in the source rest frame is between 60 and 100 microns."386 Therefore. as our SOs/cllipticals approximately form at 22. the corresponding maxima of emission must be redshiftecl to wavelengths greater than 180 and 300 microns respectively.," Therefore, as our S0s/ellipticals approximately form at $z > 2$, the corresponding maxima of emission must be redshifted to wavelengths greater than 180 and 300 microns respectively."387 This result. is quite robust in the sense that its qualitative features do not depend. on the cosmological parameters., This result is quite robust in the sense that its qualitative features do not depend on the cosmological parameters.388 However. quantitatively. there is a marked dillerence: the domination of late tvpe galaxies is more marked in a SCDAL model. where galaxies tend to form later on average (see Silk Devriencdt 2000).," However, quantitatively, there is a marked difference: the domination of late type galaxies is more marked in a SCDM model, where galaxies tend to form later on average (see Silk Devriendt 2000)."389 This remark also applies to the model of 600. where the phenomenologically evolving VLIRG fraction was the dominant. contributor in the far-LR (175 microns)," This remark also applies to the model of DG00, where the phenomenologically evolving ULIRG fraction was the dominant contributor in the far-IR (175 microns)."390 Although the general agreement of our predicted counts with the multi-wavelongth data seems. quite impressive. there are several caveats.," Although the general agreement of our predicted counts with the multi-wavelength data seems quite impressive, there are several caveats."391 At 15 microns. for example. one would sav that we match the integral counts fairly well (upper left panel of figure 3)).," At 15 microns, for example, one would say that we match the integral counts fairly well (upper left panel of figure \ref{figinf}) )."392 But. looking more closely. we cannot reproduce the change of slope seen in the LSOCAAL differential counts (fig. 4)).," But looking more closely, we cannot reproduce the change of slope seen in the ISOCAM differential counts (fig. \ref{figdiff}) )."393 There are at least a couple of reasons why this could. happen., There are at least a couple of reasons why this could happen.394 First. the SEDs of the ISOCAM galaxies. are cilferent in the mid-LR. from the ones used. here as a template. which are," First, the SEDs of the ISOCAM galaxies are different in the mid-IR from the ones used here as a template, which are"395are well-fit by a single power law and do not require more complex modelling.,are well-fit by a single power law and do not require more complex modelling.396 The actual transition from the high soft to the low hard state took place during revolutions 565 and 566 as observed in the radio band., The actual transition from the high soft to the low hard state took place during revolutions 565 and 566 as observed in the radio band.397" As we are missing the X-ray observation during this period, we cannot say whether the source has undergone an intermediate state as described in Szostek Zdziarski (2004)."," As we are missing the X-ray observation during this period, we cannot say whether the source has undergone an intermediate state as described in Szostek Zdziarski (2004)."398" The RXTE/ASM data show, however, that the transition has to be rather smooth, as the 2-10 keV flux decreases gradually from the ultrasoft state (/o-1oxey=8.3x107?ergcm? s!) to the low hard state (fo_10xev=2.5x107?ergcm? s~!)."," The /ASM data show, however, that the transition has to be rather smooth, as the 2–10 keV flux decreases gradually from the ultrasoft state $f_{2-10 \rm \,399 keV} = 8.3 \times 10^{-9} \rm \, erg \, cm^{-2} \, s^{-1}$ ) to the low hard state $f_{2-10 \rm \,400 keV} = 2.5 \times 10^{-9} \rm \, erg \, cm^{-2} \, s^{-1}$ )."401 Szostek Zdziarski (2004) suggest that the transition from the ultrasoft state to the low hard state is a transition in the hard X-rays from a jet-dominated phase to thermal Comptonization., Szostek Zdziarski (2004) suggest that the transition from the ultrasoft state to the low hard state is a transition in the hard X-rays from a jet-dominated phase to thermal Comptonization.402" In our observation, this would mean that we detect the emission of the jet in the ultrasoft state in the hard X-rays as indicated by EF~2."," In our observation, this would mean that we detect the emission of the jet in the ultrasoft state in the hard X-rays as indicated by $\Gamma \sim 2$."403" During the transition the jet becomes more and more diluted by the onset of the strong thermal Comptonization component, which we see in the steepening and brightening of the hard X-ray spectrum."," During the transition the jet becomes more and more diluted by the onset of the strong thermal Comptonization component, which we see in the steepening and brightening of the hard X-ray spectrum."404 Rajeev et al. (, Rajeev et al. (405"1994) interpreted the transition in terms of an increasing temperature of the black body component, which is at the same time decreasing in size and whose Comptonization region simultaneously is becoming more compact and more opaque.","1994) interpreted the transition in terms of an increasing temperature of the black body component, which is at the same time decreasing in size and whose Comptonization region simultaneously is becoming more compact and more opaque."406" Finally, the hard X-ray spectrum of the low hard state can be fit by a model of almost pure Compton reflection (Hjalmarsdotter et al."," Finally, the hard X-ray spectrum of the low hard state can be fit by a model of almost pure Compton reflection (Hjalmarsdotter et al."407" 2004, 2007)."," 2004, 2007)."408 The true nature of the compact object in Cygnus X-3 still has to be determined., The true nature of the compact object in Cygnus X-3 still has to be determined.409" Hard X-ray observations, as provided by and simultaneous radio observations, are essential in disentangling the four main components: the absorbing material detectable in soft X-rays, the thermal (blackbody) component, the jet, and the Comptonization component."," Hard X-ray observations, as provided by and simultaneous radio observations, are essential in disentangling the four main components: the absorbing material detectable in soft X-rays, the thermal (blackbody) component, the jet, and the Comptonization component."410Based upon our expectation that the accreting white clwarl is the dominant source of FUY flux in both svstems during quiescence. we carried oul a hieh gravity photosphere sviithetic spectral analvsis.,"Based upon our expectation that the accreting white dwarf is the dominant source of FUV flux in both systems during quiescence, we carried out a high gravity photosphere synthetic spectral analysis."411 The model atmosphere (IIlubeny.1938.TLUSTY). and spectrum svnthesis (Ihibeny&Lanz1995.SYNSPEC) codes and details of our 42 (47 per degree of freedom) minimization filling procedures are discussed in detail in Sionetal.(1995) and will not be repeated here.," The model atmosphere \citep[TLUSTY]{hub88} , and spectrum synthesis \citep[SYNSPEC]{hub95}412 codes and details of our $\chi^{2}_{\nu}$ $\chi^2$ per degree of freedom) minimization fitting procedures are discussed in detail in \citet{sio95}413 and will not be repeated here."414" To estimate physical parameters. we took the white dwarl photospheric temperature T,jj. Si and C abundances. and rotational velocity Όρο as [ree parameters."," To estimate physical parameters, we took the white dwarf photospheric temperature $_{eff}$, Si and C abundances, and rotational velocity $v_{rot}$ as free parameters."415 We normalize our fits to 1 solar radius and 1 kiloparsec such that the distance of a source is computed [rom d=1000(pe)x(Ragff.νο. or equivalently the scale [actor S=(4e2(4) ds the factor by which the theoretical flix (integrated over the FUSE wavelength range) has to be multiplied to equal the observed (integrated) flux.," We normalize our fits to 1 solar radius and 1 kiloparsec such that the distance of a source is computed from $d =4161000(pc)*(R_{wd}/R_{\odot})/\sqrt{S}$, or equivalently the scale factor $S417= \left( \frac{R_{wd}}{R_{\odot}} \right)^2 \left( \frac{d}{kpc}418\right)^{-2}$, is the factor by which the theoretical flux (integrated over the FUSE wavelength range) has to be multiplied to equal the observed (integrated) flux."419 In preparation for our model fitting of SS Aur. we masked the following wavelength regions where several narrow enmission-like features occur: 959.5 - 950.0A.. 072.4 - 972.6A.. 988.6 - 989.0A.. 1025.2 - 1026.0A.," In preparation for our model fitting of SS Aur, we masked the following wavelength regions where several narrow emission-like features occur: 959.5 - 950.0, 972.4 - 972.6, 988.6 - 989.0, 1025.2 - 1026.0."420. For RU Peg. we masked the following wavelength regions: «915Α.. O74 - 980Α.. 1029 - 1037.Α.. 1170Α.," For RU Peg, we masked the following wavelength regions: $<$ 915, 974 - 980, 1029 - 1037, $>$ 1170."421". We chose to vary the T,py. rotational velocity. and silicon and carbon abundances in our fitting."," We chose to vary the $_{eff}$, rotational velocity, and silicon and carbon abundances in our fitting."422" The erid of models extended over the following range of parameters: T,557/1000 Ux)= 22. 23. .... 55: Si — 0.1. 0.2. 0.5. 1.0. 2.0. 5.0: C = 0.1. 0.2. 0.5. 1.0. 2.0. 5.0: and οsin (kms !) = 100.200. 400. 600. 800."," The grid of models extended over the following range of parameters: $_{eff}/1000$ (K) = 22, 23, ..., 55; Si = 0.1, 0.2, 0.5, 1.0, 2.0, 5.0; C = 0.1, 0.2, 0.5, 1.0, 2.0, 5.0; and $v_{rot} \sin{i}$ (km $^{-1}$ ) = 100,200, 400, 600, 800."423 Since the distance d — 201 pe [rom the FGS parallax. we used this distance and the reduced 4? value to determine (he best-fitting model.," Since the distance d = 201 pc from the FGS parallax, we used this distance and the reduced $\chi^2$ value to determine the best-fitting model."424 In addition. as the WD is expected to be massive we fixed log g=9.0.," In addition, as the WD is expected to be massive we fixed log $g = 9.0$."425" For SS Aur. the fitting model [rom our X2 minimization routine has the following parameters: T,;;/1000 S (IX)= 33 +15/-7 Si=1.0 +1.0/-0.6 (mes solar. C = 0.1 +0.9/-0.1 limes solar.N = 2.0 +1.8/-0.7. 0,sin?=4004 kan s OMA scale [actor —3.82x10Ἱ"," For SS Aur, the best fitting model from our $\chi^2_{\nu}$ minimization routine has the following parameters: $_{eff}/1000$ (K) = 33 +15/-7 Si =1.0 +1.0/-0.6 times solar, C = 0.1 +0.9/-0.1 times solar, N = 2.0 +1.8/-0.7, $v_{rot}\sin{i} = 400\pm400$ km $^{-1}$, $\chi^2_{\nu}$, scale factor $= 3.82\times 10^{-4}$."426 The best-fitting modelis clisplavedin figure3., The best-fitting model is displayed in figure 3.427 This model gives a reasonable agreement wilh the FUSE continuum distribution and limes but vields a distance of 303 pe or 1.5 times the parallax value., This model gives a reasonable agreement with the FUSE continuum distribution and lines but yields a distance of 303 pc or 1.5 times the parallax value.428 Using the above parameters. the N abundance was found to have an upper limit N<& times solar.," Using the above parameters, the N abundance was found to have an upper limit $<$ 8 times solar."429 Next. we tried models of accretion disks alone from (he grid of Wade aid Hubeny (1998).," Next, we tried models of accretion disks alone from the grid of Wade and Hubeny (1998)."430 We fixed the inclination and the white dwarf mass at the published values of 41 degrees with Migο. , We fixed the inclination and the white dwarf mass at the published values of 41 degrees with $M_{wd} = 1.2 M_{\odot}$.431The resulting best-fit had the following parameters: 47 tà= 2.56. scale factor =1.15x10? and an accretion rate M=10MAL |. corresponding to a distance of 931 pc. or 4.6 times the parallax distance.," The resulting best-fit had the following parameters: $\chi^2_{\nu}$ = 2.56, scale factor $= 1.15\times 10^{-2}$ and an accretion rate $\dot{M} = 10^{-10} M_{\odot}$ $^{-1}$, corresponding to a distance of 931 pc, or 4.6 times the parallax distance."432 We conclude that an optically thick accretion disk by itself does not satisfactorily account for the FUSE spectrum., We conclude that an optically thick accretion disk by itself does not satisfactorily account for the FUSE spectrum.433 We assessed (he effectiveness of combining an aceretion disk with a white cwarl. again fixing the white dwarf mass at 1.2AZ.. the inclination angle of the disk at 41 degrees.," We assessed the effectiveness of combining an accretion disk with a white dwarf, again fixing the white dwarf mass at $1.2 M_{\odot}$ , the inclination angle of the disk at 41 degrees."434 The, The435"where EW,, is the measured EW of the system and L is the mean continuum light ratio at the both side of the C IL A 4267 line.",where $ EW_{m} $ is the measured EW of the system and $ L $ is the mean continuum light ratio at the both side of the C II $\lambda$ 4267 line.436" The corrected EWo/X values for the primary stars of the Algols are also plotted in Fig.2 versus the effective temperatures, in logarithmic scale."," The corrected $ EW_{0}/\lambda $ values for the primary stars of the Algols are also plotted in Fig.2 versus the effective temperatures, in logarithmic scale."437" Neither intermediate- nor narrow-band photometric observations exist for the system V1898 Cyg, therefore, we adopted the effective temperature for the primary star from Dervisoogllu et al. ("," Neither intermediate- nor narrow-band photometric observations exist for the system V1898 Cyg, therefore, we adopted the effective temperature for the primary star from Dervişooğllu et al. ("4382011).,2011).439 The EWs of M267 C II line in the gainers in the classical Algols are systematically smaller than that of the standard stars in the main-sequence band having similar effective temperatures., The EWs of $\lambda$ 4267 C II line in the gainers in the classical Algols are systematically smaller than that of the standard stars in the main-sequence band having similar effective temperatures.440 The weakest line is observed in the primary of AU Mon in which the EW is only 20 per cent that of the standard star with similar effective temperature., The weakest line is observed in the primary of AU Mon in which the EW is only 20 per cent that of the standard star with similar effective temperature.441" The weak line stars following the system AU Mon are GT Cep, HU Tau, U Cep and TU Mon in which the EWs are 22, 28, 34 and 40 per cent, respectively."," The weak line stars following the system AU Mon are GT Cep, HU Tau, U Cep and TU Mon in which the EWs are 22, 28, 34 and 40 per cent, respectively."442" We computed new-ODF ATLAS9 (Kurucz 1993) model atmospheres assuming a solar chemical composition ([M/H]=0.0), a microturbulent velocity £— 2 km/s, a surface gravity loggg—4 and the effective temperatures given in Tables 3 and 4 for each star."," We computed new-ODF ATLAS9 (Kurucz 1993) model atmospheres assuming a solar chemical composition ([M/H]=0.0), a microturbulent velocity $\xi$ = 2 km/s, a surface gravity $\log$ g=4 and the effective temperatures given in Tables 3 and 4 for each star."443" Using a Linux version (Castelli, 2005) of the WIDTH-code(Kurucz 1993) and the corrected EWs we determined the carbon abundances for the primaries of classical Algols in our list."," Using a Linux version (Castelli, 2005) of the WIDTH-code(Kurucz 1993) and the corrected EWs we determined the carbon abundances for the primaries of classical Algols in our list."444" We used the line data from the database with the version 4, and adopted the log gf-values of +0.562 and +0.717 for this doublet."," We used the line data from the database with the version 4, and adopted the log gf-values of +0.562 and +0.717 for this doublet."445 We added Stark damping constant log(ys/Ne) = -4.76 (Griem 1974) for all the stars., We added Stark damping constant $\log(\gamma_S/N_e)$ = -4.76 (Griem 1974) for all the stars.446 We, We447 Finally. although our model will be generic. we apply the results to the particular case of y-Cephet where we analyze how the uncertainties in the orbital fits of the secondary stellar companion may affect the evolution of a planetesimal swarm.," Finally, although our model will be generic, we apply the results to the particular case of $\gamma$ -Cephei where we analyze how the uncertainties in the orbital fits of the secondary stellar companion may affect the evolution of a planetesimal swarm."448 The paper is organized as follows., The paper is organized as follows.449 Section 2 presents our second-order perturbation model and analytical approximations for both the forced eccentricity and. secular frequency., Section 2 presents our second-order perturbation model and analytical approximations for both the forced eccentricity and secular frequency.450 Since we focus our attention on y-Cephei. Section 3 discusses the orbital parameters determined for the two stellar components and their precision.," Since we focus our attention on $\gamma$ -Cephei, Section 3 discusses the orbital parameters determined for the two stellar components and their precision."451 The secular dynamics of individual planetesimals. under the additional effects of a nonlinear gas drag. is analyzed in Section 4+.," The secular dynamics of individual planetesimals, under the additional effects of a nonlinear gas drag, is analyzed in Section 4."452 We also present an example of resonance trapping obtained with divergent migration., We also present an example of resonance trapping obtained with divergent migration.453 Finally. discussions close the paper in Section 5.," Finally, discussions close the paper in Section 5."454 Let us assume a small planetesimal of mass zi 1n circumstellar motion around a star of mass gr. which is in turn part of à binary system with a smaller component of mass sip.," Let us assume a small planetesimal of mass $m$ in circumstellar motion around a star of mass $m_A$, which is in turn part of a binary system with a smaller component of mass $m_B$."455 Let dp be the miy-centric semimajor axis of 15 and eg its orbital eccentricity., Let $a_B$ be the $m_A$ -centric semimajor axis of $m_B$ and $e_B$ its orbital eccentricity.456 We further assume that all motion oceurs in a plane., We further assume that all motion occurs in a plane.457 Neglecting the gravitational effects of 5! on both stellar bodies. the orbit of mp will be a fixed ellipse. while the motion of the small planetesimal will be perturbed by the gravitational effects stemming from the secondary component.," Neglecting the gravitational effects of $m$ on both stellar bodies, the orbit of $m_B$ will be a fixed ellipse, while the motion of the small planetesimal will be perturbed by the gravitational effects stemming from the secondary component."458 Thus. in our dynamical system. 775 will play the role of the perturber. while m Will be the perturbed mass.," Thus, in our dynamical system, $m_B$ will play the role of the perturber, while $m$ will be the perturbed mass."459 Outside any significant mean-motion resonance. the orbital evolution of ii will be dominated by the secular perturbations. and the short-period terms (associated to the mean longitudes) can be eliminated by a perturbation technique known àsa," Outside any significant mean-motion resonance, the orbital evolution of $m$ will be dominated by the secular perturbations, and the short-period terms (associated to the mean longitudes) can be eliminated by a perturbation technique known as."460"veraging, The expression for the secular disturbing function Α usually employed for these studies was originally developed by Heppenheimer (1978) which. except for constant terms. is given by [| (see Terquem and Papaloizou 2002). where G is the gravitational constant. a is the /4-centric semimajor axis of the planetesimal. e its eccentricity. and c its longitude of pericenter."," The expression for the secular disturbing function $R$ usually employed for these studies was originally developed by Heppenheimer (1978) which, except for constant terms, is given by R = ] (see Terquem and Papaloizou 2002), where ${\cal G}$ is the gravitational constant, $a$ is the $m_A$ -centric semimajor axis of the planetesimal, $e$ its eccentricity, and $\varpi$ its longitude of pericenter."461 The angle cg denotes the longitude of pericenter of the orbit of 5. assumed constant.," The angle $\varpi_B$ denotes the longitude of pericenter of the orbit of $m_B$, assumed constant."462 Expression (1)) 1s constructed from Kaula’s (1962) expansion of the disturbing potential. truncated to second-order expansion tn the eccentricity of the perturbed body. and performing a first-order “scissors” averaging (with respect to the masses) in the mean longitudes.," Expression \ref{eq1}) ) is constructed from Kaula's (1962) expansion of the disturbing potential, truncated to second-order expansion in the eccentricity of the perturbed body, and performing a first-order “scissors” averaging (with respect to the masses) in the mean longitudes."463 We refer to the resulting expressions as afirst-order model for the secular dynamics., We refer to the resulting expressions as a model for the secular dynamics.464 Since R does not depend explicitly on the mean longitude /1 of the planetesimal. its semimajor axis is constant and equal to the proper value αἲ.," Since $R$ does not depend explicitly on the mean longitude $\lambda$ of the planetesimal, its semimajor axis is constant and equal to the proper value $a^*$."465" Consequently. the secular system ts reduced to a single degree of freedom. and the differential equations governing the regular variables k=ecos(@—-wp) and /r=esin(@—cag) can be written asο where e, = tte Given arbitrary initial conditions (Ao.Πο). these equations admit periodic solutions of the form k(t) = e, + e;(5) h(t) = e,sin(gr@o).. where g is the secular frequency. e=(kyοι)”+Ir, is usually known as the proper (or free) eccentricity. and the phase angle is given by the expression tano=Που—οµ)."," Consequently, the secular system is reduced to a single degree of freedom, and the differential equations governing the regular variables $k=e \cos{(\varpi-\varpi_B)}$ and $h=e \sin{(\varpi-\varpi_B)}$ can be written as = -g h; = g (k - e_f), where g = ( )^3 e_f = e_B. Given arbitrary initial conditions $(k_0,h_0)$, these equations admit periodic solutions of the form k(t) = e_p + e_f h(t) = e_p, where $g$ is the secular frequency, $e_p^{2}=(k_0-e_f)^{2}+h_0^{2}$ is usually known as the proper (or free) eccentricity, and the phase angle is given by the expression $\tan \phi_0=h_0/(k_0-e_f)$."466 The constant term e; is known as the forced eccentricity and is only present in systems with an eccentric perturber., The constant term $e_f$ is known as the forced eccentricity and is only present in systems with an eccentric perturber.467 Adopting fixed values for ap and eg. equation (4)) implies that e; 1s a linear function ofthe proper semimajor axis (ej~a) while the secular frequency scales as g~αἱNY ," Adopting fixed values for $a_B$ and $e_B$, equation \ref{eq4}) ) implies that $e_f$ is a linear function ofthe proper semimajor axis $e_f \sim a^*$ ) while the secular frequency scales as $g \sim {a^*}^{3/2}$."468Our first task is to assess the accuracy of the secular solutions (5)) corresponding to the first-order model (1))., Our first task is to assess the accuracy of the secular solutions \ref{eq5}) ) corresponding to the first-order model \ref{eq1}) ).469 Two quantities we particularly wish to test are e; and sg., Two quantities we particularly wish to test are $e_f$ and $g$.470 The forced eccentricity is crucial in determining the equilibrium eccentricity of planetesimals under the effects of gas drag from the protoplanetary nebula., The forced eccentricity is crucial in determining the equilibrium eccentricity of planetesimals under the effects of gas drag from the protoplanetary nebula.471 Although any secular oscillatory motion ts expected to be damped in a gas-rich scenario. the magnitude of g is important for establishing the validity of the averaging process of the disturbing function.," Although any secular oscillatory motion is expected to be damped in a gas-rich scenario, the magnitude of $g$ is important for establishing the validity of the averaging process of the disturbing function."472" For our computations. we assume a generic binary system with mass ratio between the components of 515/nt4=0.4 and eccentricity ο,=0.36."," For our computations, we assume a generic binary system with mass ratio between the components of $m_B/m_A = 0.4$ and eccentricity $e_B=0.36$."473 This value is similar to the best-fit solution found by Hatzes et al. (, This value is similar to the best-fit solution found by Hatzes et al. (4742003) for y-Cepher.,2003) for $\gamma$ -Cephei.475 Figure 1. shows three different calculations of the forced eccentricity (top frame) and the secular frequency (bottom frame)., Figure \ref{fig1} shows three different calculations of the forced eccentricity (top frame) and the secular frequency (bottom frame).476 The value of e; appears to grow linearly with the proper semimajor axis. reaching values of ~0.1 for e~0.2up.," The value of $e_f$ appears to grow linearly with the proper semimajor axis, reaching values of $\sim 0.1$ for $a^* \sim 0.2 a_B$."477 The numerical calculations were obtained from a long-term integration of the exact equations of motion. after an online application of a low-pass FIR filter (e.g. Carpino et al.," The numerical calculations were obtained from a long-term integration of the exact equations of motion, after an online application of a low-pass FIR filter (e.g. Carpino et al."478 1987)., 1987).479 A digital filter is a numerical tool that eliminates certain frequencies from an input signal., A digital filter is a numerical tool that eliminates certain frequencies from an input signal.480 For example. given a certain time series (e.g. eccentricity as function of time) and a pass frequency vj. applying a low-pass filter signal will yield an output that maintains all the periodic variations with," For example, given a certain time series (e.g. eccentricity as function of time) and a pass frequency $\nu_{\rm pass}$ , applying a low-pass filter signal will yield an output that maintains all the periodic variations with"481object and instead go through stable mass transfer from the original primary.,object and instead go through stable mass transfer from the original primary.482 Some svstenis can further evolve through a CE phase aud form tightly bound TACINBs with ILe-ich donors. the relative contribution of which depends strougly on the assuued CE efiicicucy.," Some systems can further evolve through a CE phase and form tightly bound HMXBs with He-rich donors, the relative contribution of which depends strongly on the assumed CE efficiency."483" Ou the other hand. binaries that evolve through a CE phase before compact object formation ond up with tight enough orbits to survive ICC-SN events with ypical natal kicks in excess of ~LOOkkuss ο,"," On the other hand, binaries that evolve through a CE phase before compact object formation end up with tight enough orbits to survive ICC-SN events with typical natal kicks in excess of $\sim100$ $^{-1}$."484 These Mnarics tend to remain tight after the SN explosion and can become bright UAINBs even with uuevolved. MS donors.," These binaries tend to remain tight after the SN explosion and can become bright HMXBs even with unevolved, MS donors."485 These systems form a background of short ρου TIAINBs through which the ECS dump cau rie depending onu the how favorable ECS condition:4. are (progenitor nass rauge and natal kicks)., These systems form a background of short period HMXBs through which the ECS bump can rise depending on the how favorable ECS conditions are (progenitor mass range and natal kicks).486 Since re creation of post-CE ΠΑΛΙΟΣ is not dependent on i0 evolution of the donor. most MS-IIMXNDs form yonmediately after the SN of the primary (25 Alyy) uid their uuubers exponentially decay thereafter due to ubsequent Roche-lobe overflow by the MS donor aud Itimately mergers.," Since the creation of post-CE HMXBs is not dependent on the evolution of the donor, most MS-HMXBs form immediately after the SN of the primary $<25$ Myr) and their numbers exponentially decay thereafter due to subsequent Roche-lobe overflow by the MS donor and ultimately mergers."487 Lastly. we uote that the formation of svstems with wide orbital periods (ereater than ~10 davs) is highly uulikely to result from ICC-SN activity involving large SN natal kicks (1.0 NS from ποιος events and low-mass BIT from ICC superuovae aud οfallback).," Lastly, we note that the formation of systems with wide orbital periods (greater than $\sim10$ days) is highly unlikely to result from ICC-SN activity involving large SN natal kicks (i.e. NS from non-ECS events and low-mass BH from ICC supernovae and fallback)."488 Instead this population is dominated by ECS NS aud BIT formed through direct collapse. given the small kicks adopted for such eveuts.," Instead this population is dominated by ECS NS and BH formed through direct collapse, given the small kicks adopted for such events."489" Ποπονο, as shown for ECS-IINND in Fiewe Ll (top). even these small kicks are enough to disrupt the widest of IININDs."," However, as shown for ECS-HMXB in Figure \ref{ecsplot}~ (top), even these small kicks are enough to disrupt the widest of HMXBs."490 We have shown that. if NS formation through ECS events is associated with natal kicks smaller than ~5üklauss |. production of relatively bright wined-fed HMNDs(Ly —1xl07?eorgss 1) is significantly favored )etween 20 and λα after delta function star ornation.," We have shown that, if NS formation through ECS events is associated with natal kicks smaller than $\sim50$ $^{-1}$, production of relatively bright wind-fed HMXBs $_X$ $>$ 1 x $^{32}$ $^{-1}$ ) is significantly favored between 20 and Myr after delta function star formation."491 We call this increased TAINB formation. which appears above a background TAINB population ormed through standard ICC compact object formation. he ECSbienp.," We call this increased HMXB formation, which appears above a background HMXB population formed through standard ICC compact object formation, the ECS."492 The width iu post-starburst age of this ECS buup aud its amplitude relative to the rest of he ICC-IININDB. depend primarily on two ECS factors: the mass range of ECS progenitors and he typical maenitude of EC'S natal kicks (Figure 1))., The width in post-starburst age of this ECS bump and its amplitude relative to the rest of the ICC-HMXB depend primarily on two ECS factors: the mass range of ECS progenitors and the typical magnitude of ECS natal kicks (Figure \ref{ecsplot}) ).493 Moreover. we find that these ECS-TAINBs form through a specific evolutionary chanuucl that avoids ai CE xàase before the SN aud instead includes stable mass ransfer from the primary and a lass ratio dmnversionu eadiug to orbital expansion.," Moreover, we find that these ECS-HMXBs form through a specific evolutionary channel that avoids a CE phase before the SN and instead includes stable mass transfer from the primary and a mass ratio inversion leading to orbital expansion."494 As a result. ECS-ITNINDs ive predonmünautle wide orbits (~500 RR.) aud evolved donors.," As a result, ECS-HMXBs have predominantly wide orbits $\sim$ $_\odot$ ) and evolved donors."495 These characteristics. aloug with their sensitivity. to ECS properties. provide us with au intriguing opportunity to probe ECS plivsics and progenitors through studies of starbursts at different ages.," These characteristics, along with their sensitivity to ECS properties, provide us with an intriguing opportunity to probe ECS physics and progenitors through studies of starbursts at different ages."496 The SAIC is a promising candidate for a study probing ECS physics., The SMC is a promising candidate for a study probing ECS physics.497 Observations of the SMC lave revealed a dnajor burst of star formation 3060 Myr ago (??).. ," Observations of the SMC have revealed a major burst of star formation 30–60 Myr ago \citep{2001A&A...379..864M, 2004AJ....127.1531H}."498"Recent observations have found a surprisinely large yopulation of IININDs with Be donors concentrated iu he SAIC bar region (2??)., "," Recent observations have found a surprisingly large population of HMXBs with Be donors concentrated in the SMC bar region \citep{2000A&A...359..573H, 2004ApJ...609..133M, 2009ApJ...697.1695A}."499There are four reasons to prefer the lypothesis that Be IINXDs are primarily formed through ECS creation channels., There are four reasons to prefer the hypothesis that Be HMXBs are primarily formed through ECS creation channels.500" First. the population of IINUNDs in the SAIC ws been found to have wide orbital separations and ow eccentricities, which would requie fiie tuning of he initial orbital parameters if strong kicks were used (?2).."," First, the population of HMXBs in the SMC has been found to have wide orbital separations and low eccentricities, which would require fine tuning of the initial orbital parameters if strong kicks were used \citep{2002ApJ...574..364P}."501 Second. we note the coincidence of age between the oak of ECS driven WAINB activity. aud the observed age of star forming regions of the SAIC bar.," Second, we note the coincidence of age between the peak of ECS driven HMXB activity, and the observed age of star forming regions of the SMC bar."502 Third. the spatial distribution of SAIC IIMNDs is highly. peaked in the bar region. with few IINMXDs elsewhere in the ealaxy.," Third, the spatial distribution of SMC HMXBs is highly peaked in the bar region, with few HMXBs elsewhere in the galaxy."503 This spatial distribution is liehly unlikely to result from hieh velocity natal kicks ουπαπάς between 10 and LO Myr before the preseut., This spatial distribution is highly unlikely to result from high velocity natal kicks occuring between 10 and 40 Myr before the present.504 Lastly? have recently argued that Be stars form through spin up due to mass transfer in binaries that experience mass ratio iuversion and associated orbital expansion. which fits the evolution patlivav found for the ECS TAINBs in our models.," Lastly, \citet{2005ApJS..161..118M} have recently argued that Be stars form through spin up due to mass transfer in binaries that experience mass ratio inversion and associated orbital expansion, which fits the evolution pathway found for the ECS HMXBs in our models."505" Tn order to estimate the overcoucentration of ECS sources in the SAIC bar we select ai value of Ones = 26.5 lan ο, and cxamine the nuniber of QAINB with Ly Lx 10coress | between 30. an 60 Myr after star formation. taking snapshots of the svstenus position every 0.1 Myr."," In order to estimate the overconcentration of ECS sources in the SMC bar, we select a value of $\sigma_{ECS}$ = 26.5 km $^{-1}$, and examine the number of HMXB with $_X$ $>$ 1 x $^{32}$ $^{-1}$ between 30 and 60 Myr after star formation, taking snapshots of the systems position every 0.1 Myr."506 In Figure 3.. we plot the cumulative distribution function of three-cimensioua distances traveled due to the SN natal kick of both ECS and ICC sources. normalized to their total umber.," In Figure \ref{dist}, we plot the cumulative distribution function of three-dimensional distances traveled due to the SN natal kick of both ECS and ICC sources, normalized to their total number."507 We ignore auv effects due to the galactic eravitationa potential on the motion of IAINB sources., We ignore any effects due to the galactic gravitational potential on the motion of HMXB sources.508 We note that while overall there are approximately twice as many ICC sources. only around of ICC sources are fore within 1 kpc of the location in which they were formed.," We note that while overall there are approximately twice as many ICC sources, only around of ICC sources are found within 1 kpc of the location in which they were formed."509 Tuside this region. ECS sources domüuate bv a factor of approximately 1].," Inside this region, ECS sources dominate by a factor of approximately 4."510 We uote that the deep Chandra observations concentrate on the SMC bar region. while," We note that the deep $Chandra$ observations concentrate on the SMC bar region, while"511"where the contribution from the sum over the Lyman transitions is a function of z’, and is zero at z’>zmax(n=2).","where the contribution from the sum over the Lyman transitions is a function of $z'$, and is zero at $z'>z_{\rm max}(n=2)$."512" The total Lyman a background is then just the sum of the above components: In our fiducial model, we do not explicitly take into account other soft-UV sources of such as quasars, assuming that these are sub-dominant to the stellar emission."," The total Lyman $\alpha$ background is then just the sum of the above components: In our fiducial model, we do not explicitly take into account other soft-UV sources of such as quasars, assuming that these are sub-dominant to the stellar emission."513" However, our framework makes it simple to add additional source terms to the integrand of eq. (21)),"," However, our framework makes it simple to add additional source terms to the integrand of eq. \ref{eq:Jalpha_stars}) ),"514 if the user wishes to explore such scenarios (e.g. ?))., if the user wishes to explore such scenarios (e.g. \citealt{VG09}) ).515" All of the results in this section are from an L= 1 Gpc simulation, whose ICs are sampled on a 1800° grid, with the final low-resolution boxes being 300? (3.33 Mpc cells)."," All of the results in this section are from an $L=$ 1 Gpc simulation, whose ICs are sampled on a $^3$ grid, with the final low-resolution boxes being $^3$ (3.33 Mpc cells)."516" Our fiducial model below assumes f.= 0.1,¢x=10°""M5! (~ 1 X-ray photon per stellar baryon)??,, hvo=200 eV, a= 1.5, Tyirmin=10? K for all sources (X-ray, Lyman o and ionizing), C=2, Rmax=30 Mpc, Gion=81.534 and the stellar emissivity, e, of Pop II stars from ? normalized to 4400 ionizing photons per stellar baryon."," Our fiducial model below assumes $f_\ast=0.1$, $\zeta_X=10^{57} \Msun^{-1}$ $\sim$ 1 X-ray photon per stellar , $h\nu_0=200$ eV, $\alpha=1.5$ , $T_{\rm vir, min} = 10^4$ K for all sources (X-ray, Lyman $\alpha$ and ionizing), $C=2$, $R_{\rm max}=30$ Mpc, $\zeta_{\rm ion} = 31.5$ and the stellar emissivity, $\varepsilon$, of Pop II stars from \citet{BL05_WF} normalized to 4400 ionizing photons per stellar baryon."517" The free parameters pertaining to the spin temperature evolution were chosen to match those in ? and ?,, to facilitate comparison."," The free parameters pertaining to the spin temperature evolution were chosen to match those in \citet{Furlanetto06} and \citet{PF07}, to facilitate comparison."518" It is trivial to customize the code to add, for example, redshift or halo mass dependences to these free parameters."," It is trivial to customize the code to add, for example, redshift or halo mass dependences to these free parameters."519" The impressive length of the above list of uncertain astrophysical parameters (which itself is only a simplified description of the involved processes) serves well to underscore the need for a fast, portable code, capable of quickly scrolling through parameter space."," The impressive length of the above list of uncertain astrophysical parameters (which itself is only a simplified description of the involved processes) serves well to underscore the need for a fast, portable code, capable of quickly scrolling through parameter space."520" We also note that the 7s calculations outlined in $3 are the slowest part of the 21cmFAST code (as they involve tracking evolution down to the desired redshift), and therefore should only be used in the regime where they are important (z>17 in our fiducial model)."," We also note that the $\Ts$ calculations outlined in \ref{sec:heating} are the slowest part of the 21cmFAST code (as they involve tracking evolution down to the desired redshift), and therefore should only be used in the regime where they are important $z\gsim17$ in our fiducial model)."521" For example, generating a dT; box, assuming Ts>>T, on a 300? grid takes only a few minutes on single processor (depending on the choice of higher resolution for sampling the ICs)."," For example, generating a $\delT$ box, assuming $\Ts \gg \Tcmb$, on a $^3$ grid takes only a few minutes on single processor (depending on the choice of higher resolution for sampling the ICs)."522" However, including the spin temperature field takes an additional day of computing time."," However, including the spin temperature field takes an additional day of computing time."523" Nevertheless, once the spin temperature evolution is computed for a given realization at z, all of the intermediate outputs at z’>z can be used to compute OT; at those redshifts at no additional computation cost."," Nevertheless, once the spin temperature evolution is computed for a given realization at $z$, all of the intermediate outputs at $z' > z$ can be used to compute $\delT$ at those redshifts at no additional computation cost."524" Before showing detailed results, it would be useful to summarize the various evolutionary stages (c.f."," Before showing detailed results, it would be useful to summarize the various evolutionary stages (c.f."525 $3.1 in ?))., 3.1 in \citealt{PF07}) ).526 The reader is encouraged to reference theevolutionof the mean temperatures shown in Fig., The reader is encouraged to reference theevolutionof the mean temperatures shown in Fig.527" 10 and/or view the full movie available at http://www.astro.princeton.edu/-mesinger/Movies/delT.mov, while reading below."," \ref{fig:early_global_evolution} and/or view the full movie available at $\sim$ mesinger/Movies/delT.mov, while reading below."528differences in these two studies.,differences in these two studies.529 However the precise cause of this clisparity is worthy of further investigation. especially as it has strong implications for the design of future combined optical and racio weak lensing studies.," However the precise cause of this disparity is worthy of further investigation, especially as it has strong implications for the design of future combined optical and radio weak lensing studies."530 In this section. we describe a supplementary approach to radio shear measurements., In this section we describe a supplementary approach to radio shear measurements.531 We are motivated by Muxlowetal.(2007) who [find 92 radio sources at a detection hresholdl of 405]y within a LOO10 region centered on the IIDE-N. This should be compared with the size of the LIST ACS z-band catalogue. which contains 213000 galaxies xiehter than m.=28.3 in the same region.," We are motivated by \citet{2007ASPC..380..199M} who find 92 radio sources at a detection threshold of $\mu$ Jy within a $10^{\prime}\times 10^{\prime}$ region centered on the HDF-N. This should be compared with the size of the HST ACS $z$ -band catalogue, which contains $\simeq13000$ galaxies brighter than $m_{z}=28.3$ in the same region."532 Although a vast majority of these sources are not detected individually at radio wavelengths. it is possible to statistically detect hese verv weak radio sources.," Although a vast majority of these sources are not detected individually at radio wavelengths, it is possible to statistically detect these very weak radio sources."533 Figure 15. shows the mean racio flux measured. within a 0.75 arcsecond. radius of the »ositions of 25000 of these sources. as a function of their optical magnitude.," Figure \ref{fig:statdet} shows the mean radio flux measured within a 0.75 arcsecond radius of the positions of $\simeq8000$ of these sources, as a function of their optical magnitude."534 Radio emission at the level of a few μον ds statistically detected: from. optical svstems as faint as m.= 25: this is in good agreement with the analvsis conducted by Muxlowetal.(2007).., Radio emission at the level of a few $\mu$ Jy is statistically detected from optical systems as faint as $m_{z}=25$ ; this is in good agreement with the analysis conducted by \citet{2007ASPC..380..199M}.535 The lower data points in Figure 15. show mean flux when aperture positions are chosen at random., The lower data points in Figure \ref{fig:statdet} show mean flux when aperture positions are chosen at random.536 Since the surface. brightness of these objects can be statistically detected. we can also attempt to statistically quantify thei ellipticities.," Since the surface brightness of these objects can be statistically detected, we can also attempt to statistically quantify their ellipticities."537 In. order to do so. we choose optical ealaxies in a particular magnitude bin. rotate them so that their major axes are parallel. and create a composite optical galaxy by finding the median. value of all galaxy surface brightnesses on a pixel by pixel basis.," In order to do so, we choose optical galaxies in a particular magnitude bin, rotate them so that their major axes are parallel, and create a composite optical galaxy by finding the median value of all galaxy surface brightnesses on a pixel by pixel basis."538" We then use the positions of the optical galaxies to extract. postage- ""s .; ; ⋅⋪ ⇁ − these⊳∖⋯⊔↓↓≻⊳∖⇂↓∪⊔↓⇂↓↕∢⋅↓⋯∐∪↓⊔↓⋜↧", We then use the positions of the optical galaxies to extract postage-stamps from the radio image of diameter $''$.539⋏∙≟∢⊾∪⇂∠⊔⋜⋯↓⋖⊾↿∢⋅↓⇀≻⊳⋀⋅≻⊳∖∖⋖⋅↓∪↿⋜⋯⋅ by the same amount as the relevant optical image. aud median average these also.," We rotate these by the same amount as the relevant optical image, and median average these also."540 Phe resulting stacked optical and radio images are shown in Figure 16.., The resulting stacked optical and radio images are shown in Figure \ref{fig:blinde}. .541 As expected the stacked: rotated: optical objects are elliptical in shape., As expected the stacked rotated optical objects are elliptical in shape.542 Lf radio object ellipticities were well-aligned. with optical ellipticities. we would see the stacked rotated: radio objects having an ellipticitv on our plots: instead. they are rather circular in shape. suggesting a low correlation in alignment. in confirmation of our result in refeross..," If radio object ellipticities were well-aligned with optical ellipticities, we would see the stacked rotated radio objects having an ellipticity on our plots; instead, they are rather circular in shape, suggesting a low correlation in alignment, in confirmation of our result in \\ref{cross}. ."543 Quantitativelv. we measure the ellipticitv of the images in Figure 16..," Quantitatively, we measure the ellipticity of the images in Figure \ref{fig:blinde}."544 We caleulate the quadrupole moments of the images. where £07) is the intensity. profile of the object. and WE) is a weighting function. which here is a top-hat function with radius equal to a third. of the radius of the postage stamp.," We calculate the quadrupole moments of the images, where $I(\vec{x})$ is the intensity profile of the object and $W(\vec{x})$ is a weighting function, which here is a top-hat function with radius equal to a third of the radius of the postage stamp."545 The 2-component cllipticity can then be calculated as The calculated. ellipticities are shown as a function of the + band magnitude in Figure 17.. with errors calculated via jack-knifing of the object ellipticities in cach magnitude bin.," The 2-component ellipticity can then be calculated as The calculated ellipticities are shown as a function of the $z-$ band magnitude in Figure \ref{fig:blindeplot}, with errors calculated via jack-knifing of the object ellipticities in each magnitude bin."546 We see that the ellipticities of the rotated stacked racio images are much lower than the optical stacked: images: again. this acts as a confirmation of our results in refeross.. as it means that radio emission is not coherently oriented with respect to the optical emission.," We see that the ellipticities of the rotated stacked radio images are much lower than the optical stacked images; again, this acts as a confirmation of our results in \\ref{cross}, as it means that radio emission is not coherently oriented with respect to the optical emission."547 We see that at optical magnitudes ereater than m.=23 the error on the radio ellipticitv is substantial. as the radio emission is not well characterized bevond this magnitude limit (see bottom. panel of Figure 162).," We see that at optical magnitudes greater than $m_{z}=23$ the error on the radio ellipticity is substantial, as the radio emission is not well characterized beyond this magnitude limit (see bottom panel of Figure \ref{fig:blinde}) )."548 We have therefore applied a magnitude cut of ms.s23 in what follows., We have therefore applied a magnitude cut of $m_{z}\leq23$ in what follows.549 The natural extension to this analysis is to form blind radio shear correlation functions., The natural extension to this analysis is to form blind radio shear correlation functions.550 By using only the positions of the optical objects we can extract postage stamps from the radio image ancl again calculate the ellipticites using the quadrupole approach above: we can then calculate the correlation functions for the simple shear estimator &ες. 4623)., By using only the positions of the optical objects we can extract postage stamps from the radio image and again calculate the ellipticites using the quadrupole approach above; we can then calculate the correlation functions for the simple shear estimator $\gamma\simeq\epsilon/(2-\langle\epsilon^2\rangle)$ .551" ""Phe resulting correlation functions are shown as solid lines in Figure 18..", The resulting correlation functions are shown as solid lines in Figure \ref{fig:blindcfn}.552 Phe errors from this technique are comparable with those found for optical-xradio correlations in section 5.4: resulting cosmological constraints are shown in Figure 19.. amounting to We can also make use of the optical shear estimators we have already calculated.," The errors from this technique are comparable with those found for optical-radio correlations in section \ref{cosmic}; resulting cosmological constraints are shown in Figure \ref{fig:blindcosmo}, amounting to We can also make use of the optical shear estimators we have already calculated."553 We erossecorrelate these optical shears with the blind radio ellipticities. showing the resulting correlation functions as the dashed lines in Figure 18. and cosmological constraints in Figure 19..giving The fact that these constraints are comparable with those from the more conventional raclio-optical correlation functionin 54 willlead us to pursue both techniques in future work.," We cross-correlate these optical shears with the blind radio ellipticities, showing the resulting correlation functions as the dashed lines in Figure \ref{fig:blindcfn} and cosmological constraints in Figure \ref{fig:blindcosmo}, ,giving The fact that these constraints are comparable with those from the more conventional radio-optical correlation functionin \ref{cosmic} willlead us to pursue both techniques in future work."554All Herschel--PACS maps detect circular rings of emission around the three central stars.,All -PACS maps detect circular rings of emission around the three central stars.555 These detached shells suggest that the stars underwent an increase in mass-loss for a short period of time., These detached shells suggest that the stars underwent an increase in mass-loss for a short period of time.556 In all cases. we have found that varying dust emission continues to the inner parts of the detached shells.," In all cases, we have found that varying dust emission continues to the inner parts of the detached shells."557 This can be seen from the intensity profiles in Fig., This can be seen from the intensity profiles in Fig.558 2 anc the models., \ref{profiles} and the models.559 The corresponding timescales are of the order of thousands of years. whereas the age of the outer detachec material is some 1000 or even 100000 years.," The corresponding timescales are of the order of thousands of years, whereas the age of the outer detached material is some 1000 or even 000 years."560 The correlatior with the molecular observations is striking in some cases evel down to fine details. e.g.. the lower shell density of TT Cyg to the north.," The correlation with the molecular observations is striking in some cases even down to fine details, e.g., the lower shell density of TT Cyg to the north."561 More detailed modelling combining all information about the objects. additional observations. and comparison with future observations of similar objects will definitely provide a clearer understanding of the time evolution of the mass-loss process.," More detailed modelling combining all information about the objects, additional observations, and comparison with future observations of similar objects will definitely provide a clearer understanding of the time evolution of the mass-loss process."562bandpass roughly matches the intersection of the red and white spectra.,bandpass roughly matches the intersection of the red and white spectra.563" In figure 4 the intersection frequency is 1/180 days, so τε should be z180/27."," In figure 4 the intersection frequency is 1/180 days, so $\tau_s$ should be $\approx 180/2 \pi$."564 The results of this convolution are sampled at the original sampling times., The results of this convolution are sampled at the original sampling times.565 'They are then interpolated onto a regular grid using a cubic spline constrained so its step response does not overshoot (Fritsch Carlson 1980))., They are then interpolated onto a regular grid using a cubic spline constrained so its step response does not overshoot (Fritsch Carlson ).566" The white component is found by subtracting the red component, evaluated at the original sample times, from the original residuals."," The white component is found by subtracting the red component, evaluated at the original sample times, from the original residuals."567 Further details of this process are given on the web page., Further details of this process are given on the web page.568 The performance of the various fitting algorithms can be compared by simulation of observations of a pulsar with known parameters and added noise with known statistics., The performance of the various fitting algorithms can be compared by simulation of observations of a pulsar with known parameters and added noise with known statistics.569" We expected the various algorithms to be unbiased because least-squares algorithms perform well in this respect, but we found a serious bias in the algorithm which we will discuss in section 5.1."," We expected the various algorithms to be unbiased because least-squares algorithms perform well in this respect, but we found a serious bias in the algorithm which we will discuss in section 5.1."570 The primary performance measure is the rms variation in the parameter estimates found by repeating the same simulation many times., The primary performance measure is the rms variation in the parameter estimates found by repeating the same simulation many times.571 It is also important that the algorithm return estimate of the uncertainties in the parameters which agree well with the actual rms variations., It is also important that the algorithm return estimate of the uncertainties in the parameters which agree well with the actual rms variations.572 The parameters of the timing model have different effects on the residuals., The parameters of the timing model have different effects on the residuals.573 Some of them are essentially time harmonic: the position and proper motion parameters adjust the amplitude and phase of an annual sine wave; the parallax does the same for a biannual sine wave; and the binary parameters adjust sine waves at harmonics of the binary period., Some of them are essentially time harmonic: the position and proper motion parameters adjust the amplitude and phase of an annual sine wave; the parallax does the same for a biannual sine wave; and the binary parameters adjust sine waves at harmonics of the binary period.574 The spin frequency parameters v and v adjust the linear and quadratic polynomial coefficients and their effects are confined to frequencies f<1/To»vs., The spin frequency parameters $\nu$ and $\dot{\nu}$ adjust the linear and quadratic polynomial coefficients and their effects are confined to frequencies $f \le 1/T_{obs}$.575 Jumps are Heaviside step functions having a power spectrum of the form 1/f?., Jumps are Heaviside step functions having a power spectrum of the form $1/f^2$.576 Thus it is difficult to remove the timing noise using the polynomial or Fourier schemes without distorting the jump., Thus it is difficult to remove the timing noise using the polynomial or Fourier schemes without distorting the jump.577 We simulated a four-dimensional test matrix: different algorithms; different sampling; different timing noise; and different parameter types., We simulated a four-dimensional test matrix: different algorithms; different sampling; different timing noise; and different parameter types.578" We tested four algorithms: WLS, Cholesky, polynomial, and Fourier; two sampling schemes, regular and irregular; two noise types, weak red and strong red; and three parameter types, time-harmonic, broadband, and polynomial."," We tested four algorithms: WLS, Cholesky, polynomial, and Fourier; two sampling schemes, regular and irregular; two noise types, weak red and strong red; and three parameter types, time-harmonic, broadband, and polynomial."579" The weak red noise was simulated with an amplitude of A=1x10774 yy?, a corner frequency of f,=0.3yy ! and a spectral exponent a=2.5."," The weak red noise was simulated with an amplitude of $A = 1\times10^{-24}$ $^{3}$, a corner frequency of $f_c = 0.3$ $^{-1}$ and a spectral exponent $\alpha = 2.5$."580" The strong red noise had A=1x1077 yy’, fe= 0.0lyy~' and a=5.5."," The strong red noise had $A = 1\times10^{-17}$ $^3$, $f_c = 0.01$ $^{-1}$ and $\alpha = 5.5$."581 The irregular sampling was taken from actual observations of PSR J0711—6830 that contains 225 points over yy. The regular sampling had the same number of points sampled over the same data span., The irregular sampling was taken from actual observations of PSR $-$ 6830 that contains 225 points over y. The regular sampling had the same number of points sampled over the same data span.582" Each case was simulated 100 times with the same parameters, but different realizations of the red and white noise."," Each case was simulated 100 times with the same parameters, but different realizations of the red and white noise."583" For each realization we fitted for the standard pulsar timing model parameters and for comparison recorded the pulsar's right ascension, o, proper motion in right ascension, La, parallax, 7, and the size of the jump."," For each realization we fitted for the standard pulsar timing model parameters and for comparison recorded the pulsar's right ascension, $\alpha$, proper motion in right ascension, $\mu_\alpha$, parallax, $\pi$, and the size of the jump."584" We found that the algorithm was significantly biased, in the sense that the parameter estimates depended on the initial conditions."," We found that the algorithm was significantly biased, in the sense that the parameter estimates depended on the initial conditions."585" Parameters were biased towards zero, ie., towards their initial conditions."," Parameters were biased towards zero, i.e., towards their initial conditions."586 So we ran special simulations to test for initial-condition bias for a harmonic parameter (proper motion in right ascension) and a broadband parameter (a phase jump)., So we ran special simulations to test for initial-condition bias for a harmonic parameter (proper motion in right ascension) and a broadband parameter (a phase jump).587 We ran 100 simulations of the same fit with slightly different initial conditions., We ran 100 simulations of the same fit with slightly different initial conditions.588" This showed that the WLS, Cholesky and polynomial algorithms were unbiased."," This showed that the WLS, Cholesky and polynomial algorithms were unbiased."589 We compare the results for the Cholesky and algorithms in Figure 5., We compare the results for the Cholesky and algorithms in Figure 5.590 In the top panel (a) the results for a phase jump are overplotted., In the top panel (a) the results for a phase jump are overplotted.591 In this case the result has very small error bars but it is correlated with the initial condition., In this case the result has very small error bars but it is correlated with the initial condition.592 In the second and third panels (b) and (c) the results for proper motion in « are shown., In the second and third panels (b) and (c) the results for proper motion in $\alpha$ are shown.593" Here one can see that the error bars for the results are half those for the Cholesky results, but the results are heavily"," Here one can see that the error bars for the results are half those for the Cholesky results, but the results are heavily"594luminosity of this component is less than of that of the whole nebula.,luminosity of this component is less than of that of the whole nebula.595 The cutoff in the synchrotron spectrum occurs at a characteristic frequency Vpeak~4.2x106B4?.," The cutoff in the synchrotron spectrum occurs at a characteristic frequency $\nu_{\rm peak} \sim 4.2 \times 10^{6} \, B \, \gamma^{2}$."596" Provided that synchrotron radiation is the dominant mechanism through which particles channel their energy, the maximum electron Lorentz factor obtained by equating tsync tO taccel iS YmaxΟΚ(B9)13, where η>1 is the gyrofactor that characterizes the acceleration rate Vaccel=Y/taccer ANd face=mE/qeBc."," Provided that synchrotron radiation is the dominant mechanism through which particles channel their energy, the maximum electron Lorentz factor obtained by equating $t_{\rm sync}$ to $t_{\rm accel}$ is $\gamma_{\rm max} \propto (B \, \eta)^{-1/2}$, where $\eta \geq 1$ is the gyrofactor that characterizes the acceleration rate $\dot{\gamma}_{\rm accel} \equiv \gamma/t_{\rm accel}$ and $t_{\rm accel} = \eta \, E/q_{\rm e}\, B\,c$."597" This makes Vpeax independent of B, leading to an electron synchrotron energy cutoff ~16057! MeV (seee.g.?).."," This makes $\nu_{\rm peak}$ independent of $B$ , leading to an electron synchrotron energy cutoff $\approx 160 \eta^{-1}$ MeV \citep[see e.g.][]{2000NewA....5..377A}."598" A higher value may imply that the conditions in the accelerator differ from those in the emission region, e.g. there is a lower magnetic field in the former, or that the synchrotron gamma-rays are produced in a relativistically moving region, which produces a shift in the energy cutoff to higher energies by the corresponding Doppler factor ó."," A higher value may imply that the conditions in the accelerator differ from those in the emission region, e.g. there is a lower magnetic field in the former, or that the synchrotron gamma-rays are produced in a relativistically moving region, which produces a shift in the energy cutoff to higher energies by the corresponding Doppler factor $\delta$."599" In this scenario, a value 6~367/160z2.3 would be required to explain the energy cutoff obtained during the flaring episode."," In this scenario, a value $\delta \sim 367/160 \approx 2.3$ would be required to explain the energy cutoff obtained during the flaring episode."600" On the other hand, magnetic fields at the level of between ~ 300 LG and ~ 2 mG are found in the synchrotron nebula and wisps, respectively (seee.g.?).."," On the other hand, magnetic fields at the level of between $\sim$ 300 $\mu$ G and $\sim$ 2 mG are found in the synchrotron nebula and wisps, respectively \citep[see e.g.][]{2008ARA&A..46..127H}."601 Synchrotron radiation at ~ 1 GeV implies that y~3—10x10? in the emitting regions., Synchrotron radiation at $\sim$ 1 GeV implies that $\gamma \sim 3-10 \times 10^{9}$ in the emitting regions.602" Taking 5~2.3, the comoving cooling timescale for those particles, taking an extreme value B~ 2 mG, is ~0.3 d. The corresponding observer timescale would then be similar to the decay time of the peaks present in the lightcurve during the flaring period, <1 d. In contrast, the flares could be related to an enhanced electron population, and the spectral variability could be obtained by raising the continuum normalization by a factor of ~5 or by adding a hard very high energy electron population (leading to a photon index I'«1)."," Taking $\delta \sim 2.3$, the comoving cooling timescale for those particles, taking an extreme value $B \sim$ 2 mG, is $\sim 0.3$ d. The corresponding observer timescale would then be similar to the decay time of the peaks present in the lightcurve during the flaring period, $\lesssim 1$ d. In contrast, the flares could be related to an enhanced electron population, and the spectral variability could be obtained by raising the continuum normalization by a factor of $\sim 5$ or by adding a hard very high energy electron population (leading to a photon index $\Gamma<1$ )."603" In this case, a Doppler boosting may not be required, and the observed duration of the flares could correspond to the synchrotron timescale of PeV electrons embedded in magnetic fields S1mG."," In this case, a Doppler boosting may not be required, and the observed duration of the flares could correspond to the synchrotron timescale of PeV electrons embedded in magnetic fields $\lesssim 1\,\mathrm{mG}$."604 'The duration of the three short flares limits the size of the emitting region(s) to <10? cm., The duration of the three short flares limits the size of the emitting region(s) to $\lesssim 10^{15}$ cm.605" The peak luminosity of these flares is higher/brighter than 1035 erg/s, i.e. >0.5 ,oof the spin-down luminosity, assuming an isotropic distribution."," The peak luminosity of these flares is higher/brighter than $10^{35}$ erg/s, i.e. $\geqslant 0.5$ of the spin-down luminosity, assuming an isotropic distribution."606" The distance between the emitting region and the pulsar can thus be constrainedto be <6x1016 (?).. (seee.g.?),,"," The distance between the emitting region and the pulsar can thus be constrainedto be $\leqslant 6\times 10^{16}$ \citep{2010ATel.2903....1C}. \citep[see e.g.][]{2004ApJ...601..479N},"607to such interactions as galaxy mergers.,to such interactions as galaxy mergers.608" Here, we focus on those dynamical processes in non-lossils which result in the dispersion of the primordial stellar populations of the brightest satellites, the net effect of which is to either make the non-fossil populations invisible to current surveys by reducing their surface brightnesses below the SDSS detection limits or preferentially stripping them during interacüons with more massive halos."," Here, we focus on those dynamical processes in non-fossils which result in the dispersion of the primordial stellar populations of the brightest satellites, the net effect of which is to either make the non-fossil populations invisible to current surveys by reducing their surface brightnesses below the SDSS detection limits or preferentially stripping them during interactions with more massive halos."609 The former mechanism would be relevant to non-fossils at/?z;500 kpe where udal forces are negligible., The former mechanism would be relevant to non-fossils at$R \simgt 500$ kpc where tidal forces are negligible.610 The number of pre-reionization halos in à ;=0 dwarf increases with mass., The number of pre-reionization halos in a $z=0$ dwarf increases with mass.611" In this section, we explore the role of mergers to rend invisible, or strip. the primordial populations of stars in the the more massive dwarls (non-fossils)."," In this section, we explore the role of mergers to rend invisible, or strip, the primordial populations of stars in the the more massive dwarfs (non-fossils)."612" Unlike in the previous sections, here we dillerentiate between the non-fossils and polluted fossils in our simulations."," Unlike in the previous sections, here we differentiate between the non-fossils and polluted fossils in our simulations."613" We remind the reader, that though both populations have ων which were large enough for them to accrete gas [rom the [IGM in the past, only the non-fossils are at or above that threshold at z=0."," We remind the reader, that though both populations have $v_{max}$ which were large enough for them to accrete gas from the IGM in the past, only the non-fossils are at or above that threshold at $z=0$."614" When à system undergoes a galaxy merger, kinetic energy [rom the collision is imparted to the stars."," When a system undergoes a galaxy merger, kinetic energy from the collision is imparted to the stars."615" Immediately alter the collision, the new system will be in its most diffuse state."," Immediately after the collision, the new system will be in its most diffuse state."616" We define a galaxy merger as the interaction of two or more pre-reionization halos, both containing a primordial stellar population."," We define a galaxy merger as the interaction of two or more pre-reionization halos, both containing a primordial stellar population."617" Although there is significant scatter in the luminosities of minihalos of the same mass, in general, the luminous minihalos are more massive than those which are dark."," Although there is significant scatter in the luminosities of minihalos of the same mass, in general, the luminous minihalos are more massive than those which are dark."618 An interaction between two luminous minihalos is therefore more significant., An interaction between two luminous minihalos is therefore more significant.619" We use ως the number of luminous pre-reionization halos within al 2=ϐ) halo, as a proxy for the number of galaxy mergers."," We use $N_{lum}$, the number of luminous pre-reionization halos within at $z=0$ halo, as a proxy for the number of galaxy mergers."620" If there are multiple galaxy mergers in a short amount of time, these stars will be suscepuble to stripping."," If there are multiple galaxy mergers in a short amount of time, these stars will be susceptible to stripping."621" In addition, recent work on increasing the extent of the stellar population in bright ellipucals Irom z;=2 to 2=0 suggests thal many minor interactions over several Gyrs can increase the size of the galaxy by a factor of 2-5 without significantly increasing mass (?).."," In addition, recent work on increasing the extent of the stellar population in bright ellipticals from $z=2$ to $z=0$ suggests that many minor interactions over several Gyrs can increase the size of the galaxy by a factor of 2-5 without significantly increasing mass \citep{Naabetal:09}."622" Roughly, the larger the number of significant interactions, the greater the spatial extent of the pre-reionization population and more likely the halo will have lost a significant fraction of its primordial population to dynamical heating."," Roughly, the larger the number of significant interactions, the greater the spatial extent of the pre-reionization population and more likely the halo will have lost a significant fraction of its primordial population to dynamical heating."623" For isolated halos, with large ;V;,,,,. the radius of the primordial population increases, possibly until it fills the spatial extent of the dark matter halo."," For isolated halos, with large $N_{lum}$, the radius of the primordial population increases, possibly until it fills the spatial extent of the dark matter halo."624 Such an extended system would have extremely low surface brightness and would be susceptible to tidal stripping., Such an extended system would have extremely low surface brightness and would be susceptible to tidal stripping.625 We next look at which of our three subhalo populations has a significant number of members with Αρ>3., We next look at which of our three subhalo populations has a significant number of members with $N_{lum}>3$.626" The Iractüions of the true fossil, and populations with a given .Vj,,,, are shown in the left and right panels of Figure 14.."," The fractions of the true fossil, and non-fossils populations with a given $N_{lum}$ are shown in the left and right panels of Figure \ref{Nlum.hist}."627" Neither of the fossil populations has a significant fraction of subhalos with η>3 with the fractions at 15€ and ~10% respectively,", Neither of the fossil populations has a significant fraction of subhalos with $N_{lum}>3$ with the fractions at $\sim1\%$ and $\sim10\%$ respectively.628" We therefore assume, that, while a few of our true and polluted fossils may have had their primordial populations diffused by mergers, the vast majority remain dynamically cold."," We therefore assume, that, while a few of our true and polluted fossils may have had their primordial populations diffused by mergers, the vast majority remain dynamically cold."629 The non-fossils show the opposite trend., The non-fossils show the opposite trend.630" The right panel of Figure 14. shows that «10% of the have Nya,«3 and the distribution peaks at Nj, 5.", The right panel of Figure \ref{Nlum.hist} shows that $<10\%$ of the non-fossils have $N_{lum}<3$ and the distribution peaks at $N_{lum} \sim 5$ .631 A population of non-fossils would be much more likely to have a primordial population dispersed by multiple major interactions than their fossil counterparts., A population of non-fossils would be much more likely to have a primordial population dispersed by multiple major interactions than their fossil counterparts.632the drop of low excitation. lines between the adjacent knots C and D (cf.,the drop of low excitation lines between the adjacent knots C and D (cf.633 Sect. 3.2) , Sect. \ref{spatial_line_distribution}) )634requires matter bounded clouds cut at about « the HII Strómmgren radius. the exact position of the cut depending on the assumed shape of the AG continuum.," requires matter bounded clouds cut at about $\times$ the HII Strömmgren radius, the exact position of the cut depending on the assumed shape of the AGN continuum."635 Another parameter affecting the ratio of low-to-high excitation lines(e.g. [OIJ/[OIIL[) is the tron gas phase abundance which influences the cooling of the partially tonizec region (cf., Another parameter affecting the ratio of low--to--high excitation lines (e.g. [OI]/[OIII]) is the iron gas phase abundance which influences the cooling of the partially ionized region (cf.636 Sect. [ 4.2.4))., Sect. \ref{role_of_dust}) ).637 If iron is more abundant in knot D. as by its stronger |FeVII] line emission (cf.," If iron is more abundant in knot D, as indicated by its stronger [FeVII] line emission (cf."638" Table 2. and Sect. indicated . """," Table \ref{tab_obs2} and Sect. \ref{spatial_line_distribution}) ),"639athan [OI]. [SH]. [NI] could be depressed by the increased cooling.," than [OI], [SII], [NII] could be depressed by the increased [FeII] cooling."640 The abundances derived for the LINER-like knots (H. D are very uncertain (1.3 dex at least).," The abundances derived for the LINER–like knots (H, I) are very uncertain $\pm$ 1.3 dex at least)."641 Their spectra are not compatible with illumination from the same continuum seen by knot C but require a harder (1.6. more X rays relative to 13— eV photons) spectrum which could be in principle obtained by filtering the AGN continuum through an absorber with a carefully tuned photoelectric opacity., Their spectra are not compatible with illumination from the same continuum seen by knot C but require a harder (i.e. more X rays relative to 13--80 eV photons) spectrum which could be in principle obtained by filtering the AGN continuum through an absorber with a carefully tuned photoelectric opacity.642 Alternatively. the weak (in surface brightness) spectrum of these low excitation knots could be explained by shock excitation. in which case one expects j~I and a factor>» ]O larger than in the case of pure photoionization.," Alternatively, the weak (in surface brightness) spectrum of these low excitation knots could be explained by shock excitation, in which case one expects $\simeq$ 1 and a factor $>$ 10 larger than in the case of pure photoionization."643 Finally. the oxygen lines in the highly reddened HI[-like knots (E. L) are too weak to allow any reliable abundance analysis.," Finally, the oxygen lines in the highly reddened HII–like knots (E, L) are too weak to allow any reliable abundance analysis."644 Deriving abundances of AGN clouds from phototonization models has generally been considered to be unreliable because the shape of the AGN ionizing continuum and the density distribution of the emitting regions (clouds) are both basically unknown., Deriving abundances of AGN clouds from photoionization models has generally been considered to be unreliable because the shape of the AGN ionizing continuum and the density distribution of the emitting regions (clouds) are both basically unknown.645 The more typical approach therefore has been to assume a metallicity and use photoionization models to constrain the AGN spectrum and/or the gas density distribution. or just to demonstrate that the gas 1s phototonized.," The more typical approach therefore has been to assume a metallicity and use photoionization models to constrain the AGN spectrum and/or the gas density distribution, or just to demonstrate that the gas is photoionized."646" Although explicit statements concerning the nitrogen abundance are often found im the literature (e.g. Storchi-Bergmann Pastoriza 1989., Simpson Ward 1996)) these are based on a very limited choice of photoionization model parameters and in most cases find oxygen abundances close to solar. in disagreement with what is derived here."," Although explicit statements concerning the nitrogen abundance are often found in the literature (e.g. Storchi-Bergmann Pastoriza \cite{storchi89}, Simpson Ward \cite{simpson96}) ) these are based on a very limited choice of photoionization model parameters and in most cases find oxygen abundances close to solar, in disagreement with what is derived here."647 The only other piece of work which covers a model parameter range comparable to that presented here is that by Komossa Schulz (1997)) who analyzes a much more limited numbers of lines. e.g. do not include [ArIV.V]. in a large sample of Seyferts.," The only other piece of work which covers a model parameter range comparable to that presented here is that by Komossa Schulz \cite{komossa97}) ) who analyzes a much more limited numbers of lines, e.g. do not include [ArIV,V], in a large sample of Seyferts."648 They find that. on average. oxygen is underabundant by a factor of —2 and that the N/O ratio is only a factor of 1.5-2.0 above the solar value.," They find that, on average, oxygen is underabundant by a factor of $\sim$ 2 and that the N/O ratio is only a factor of 1.5–2.0 above the solar value."649 We believe that the analysis presented here leads to more reliable metallicity estimates., We believe that the analysis presented here leads to more reliable metallicity estimates.650 The results presented here indicate that. in spite of the above uncertainties. reltable metallicities can indeed be derived from spectra including a large enough number of lines.," The results presented here indicate that, in spite of the above uncertainties, reliable metallicities can indeed be derived from spectra including a large enough number of lines,"651would (hen expect the actual curve of ¢ vs 42 lo evolve to be higher than the value Eq.(19) predicts in situations with low /? values.,would then expect the actual curve of $\zeta$ vs $R$ to evolve to be higher than the value Eq.(19) predicts in situations with low $R$ values.652 Meanwhile. lor high. A. (he analvGcal prediction and the real physical outcome should both. approach the horizontal line ¢=1. which denotes conductive efficiency. consistent wilh the unmagnetized case.," Meanwhile, for high $R$, the analytical prediction and the real physical outcome should both approach the horizontal line $\zeta = 1$, which denotes conductive efficiency consistent with the unmagnetized case."653 We emphasize that H as used in this paper is alwavs calculated with the the initial values of (he magnetic liekl. not time evolved values. and that Eq.(19) is valid when estimating a cold to hot interlace with initial tangle measure as the ratio of initial global straight field to initially local tangled field.," We emphasize that $R$ as used in this paper is always calculated with the the initial values of the magnetic field, not time evolved values, and that Eq.(19) is valid when estimating a cold to hot interface with initial tangle measure as the ratio of initial global straight field to initially local tangled field."654 To follow a measure of the tangle that evolves with time. a generalized tangle measure should be calculated in a more sophisticated manner and the integral form (Eq.(16)) should be applied.," To follow a measure of the tangle that evolves with time, a generalized tangle measure should be calculated in a more sophisticated manner and the integral form (Eq.(16)) should be applied."655We choose initial conditions with values 22=0.0.0.2.0.4.0.6.1.0.2.0.4.0 to run the simulations.,"We choose initial conditions with values $R = 0.0,\,0.2,\,0.4,\,0.6,\,1.0,\,2.0,\,4.0$ to run the simulations."656 The simulation run Gime is taken to be 1.2 (which corresponds to 12.000 vears in real units for WBB.," The simulation run time is taken to be $1.2$ (which corresponds to $12,000$ years in real units for WBB."657 The initial cuts of temperature and magnetic field lines [ου R=0.0.0.4.1.0 are shownin Figure 2((a). Figure 4((a) and Fieure 5((a) respectively.," The initial cuts of temperature and magnetic field lines for $R = 0.0,\,0.4,\,1.0$ are shownin Figure \ref{fig02}( (a), Figure \ref{fig04}( (a) and Figure \ref{fig05}( (a) respectively."658 Figure 3((a) shows the initial eut of the density distribution in the /2=0.0 run., Figure \ref{fig03}( (a) shows the initial cut of the density distribution in the $R=0.0$ run.659 We also run simulations with purely horizontal magnetic [field lines. equivalent to (he 2=2€ case. and runs with purely vertical field lines.," We also run simulations with purely horizontal magnetic field lines, equivalent to the $R =\infty$ case, and runs with purely vertical field lines."660 Frames (b) to (d) in Figure 2. to Figure 5. are from the late stages of the evolution. aud the final frames always displav the steady state ol the runs.," Frames (b) to (d) in Figure \ref{fig02} to Figure \ref{fig05} are from the late stages of the evolution, and the final frames always display the steady state of the runs."661 A steady state is facilitated bv the fact that the boundaries are kept at a fixed temperature throughout the simulations., A steady state is facilitated by the fact that the boundaries are kept at a fixed temperature throughout the simulations.662 In Figure 6.. we plot the mean cuts of the temperature Z;. obtained by averaging," In Figure \ref{fig06}, , we plot the mean cuts of the temperature $T_c$ , obtained by averaging"663Alotion in à LAEW is treated in paper 1.,Motion in a LAEW is treated in paper 1.664 Assuming an electric field Ej(X)=EjT(). the 4-velocity is u(X)=ug+Gop/)£F (x). wilh oy)25(N)9(X). up=5029 à constant of integration. and F(4)=IdxTOU).," Assuming an electric field $E_0(\chi)=E_0T(\chi)$, the 4-velocity is $u(\chi)=u_0+(\omega_E/\Omega)F(\chi)$ , with $u(\chi)=\gamma(\chi)\beta(\chi)$, $u_0=\gamma_0\beta_0$ a constant of integration, and $F(\chi)=\int_0^\chi\rmd\chi'\,T(\chi')$."665 Here we assume that the particle is highly relativistic. selling 9(\)=zl. except in evaluating Z(X). which appears in the phase in (2.1)). where we asstune 2(X)=ΕΙ—1/2470))]. with wpqim.," Here we assume that the particle is highly relativistic, setting $\beta(\chi)=\pm1$, except in evaluating $Z(\chi)$ , which appears in the phase in \ref{eta1}) ), where we assume $\beta(\chi)=\pm[1-1/2\gamma^2(\chi)]$, with $\omega_E=qE_0/m$."666 A second integration gives theorbit of a particle gives GZ(\)). right].," A second integration gives theorbit of a particle gives ), ]."667dashed line.,dashed line.668" To simulate the finite size of the dust grains, we added the contribution from diffraction using a diameter of the grains of 10, 30 and 100m in dotted, dashed and solid lines, respectively."," To simulate the finite size of the dust grains, we added the contribution from diffraction using a diameter of the grains of $10$ , $30$ and $100\,\mu$ m in dotted, dashed and solid lines, respectively."669" For the diffraction, we took simple Fraunhofer diffraction by a spherical aperture."," For the diffraction, we took simple Fraunhofer diffraction by a spherical aperture."670" To get rid of the resonance structures associated with Fraunhofer diffraction, averaging was performed over a narrow size distribution (a flat distribution from 0.8 to 1.2 times the average size)."," To get rid of the resonance structures associated with Fraunhofer diffraction, averaging was performed over a narrow size distribution (a flat distribution from 0.8 to 1.2 times the average size)."671 Grains of 100 um diameter clearly fit the best to the empirical angular scattering function (shown by the thick gray line) in agreement with Dentetal.(2000)..," Grains of $100\,\mu$ m diameter clearly fit the best to the empirical angular scattering function (shown by the thick gray line) in agreement with \citet{2000MNRAS.314..702D}."672" For the observed scattering function of the grains orbiting Fomalhaut we used the Henyey-Greenstein parameterization obtained by Kalasetal.(2005),, where we simply switched the forward and backward scattering directions, ie. switching the asymmetry parameter g from +0.2 to —0.2."," For the observed scattering function of the grains orbiting Fomalhaut we used the Henyey-Greenstein parameterization obtained by \citet{2005Natur.435.1067K}, where we simply switched the forward and backward scattering directions, i.e. switching the asymmetry parameter $g$ from $+0.2$ to $-0.2$."673 The angular scattering function obtained in this way was multiplicatively scaled to match the computations., The angular scattering function obtained in this way was multiplicatively scaled to match the computations.674 This scaling factor gives the single scattering albedo of the grains., This scaling factor gives the single scattering albedo of the grains.675" We found that when we integrated the angular scattering function resulting from the Hapke theory over the angles accessible in the Fomalhaut system (the area indicated in gray), the albedo of the grains averaged over the observable range of angles is 5%."," We found that when we integrated the angular scattering function resulting from the Hapke theory over the angles accessible in the Fomalhaut system (the area indicated in gray), the albedo of the grains averaged over the observable range of angles is $5$."676. This is consistent with the low albedo found by Kalasetal.(2005).., This is consistent with the low albedo found by \citet{2005Natur.435.1067K}.677 If we consider the assumptions on the regolith particles and the Hapke theory it is clear that the model we present is quite rough., If we consider the assumptions on the regolith particles and the Hapke theory it is clear that the model we present is quite rough.678" However, a firm outcome of the model is that for the range of scattering angles we observe the scattering is dominated by reflection rather than by diffraction."," However, a firm outcome of the model is that for the range of scattering angles we observe the scattering is dominated by reflection rather than by diffraction."679" This implies, as shown above, that the grains are larger than um in diameter."," This implies, as shown above, that the grains are larger than $\,\mu$ m in diameter."680 This outcome does not depend on the assumptions in the model and is therefore a solid conclusion., This outcome does not depend on the assumptions in the model and is therefore a solid conclusion.681 We conclude that the scattering surface in the disk around Formalhaut is dominated by grains of at least 100um in size.," We conclude that the scattering surface in the disk around Formalhaut is dominated by grains of at least $100\,\mu$ m in size."682" Our findings agree with the fact that the infrared spectrum is featureless (Stapelfeldtetal., 2004),, indicating that grains smaller than a few micron are heavily depleted in the system."," Our findings agree with the fact that the infrared spectrum is featureless \citep{2004ApJS..154..458S}, , indicating that grains smaller than a few micron are heavily depleted in the system."683" Also, Dentetal.(2000) find that large grains dominate the thermal emission."," Also, \citet{2000MNRAS.314..702D} find that large grains dominate the thermal emission."684 It is remarkable that such large grains are observed directly in optical light in an astronomical object., It is remarkable that such large grains are observed directly in optical light in an astronomical object.685" The reason for this is that while the dust mass in many systems is dominated by large grains (seee.g.Wilneretal.,2005;Testietal., 2003),, they are normally over-shone by a component of small grains."," The reason for this is that while the dust mass in many systems is dominated by large grains \citep[see e.g.][]{2005ApJ...626L.109W, 2003A&A...403..323T}, they are normally over-shone by a component of small grains."686" This component, though less massive, dominates the optical cross section, and therefore appearance, of the disk."," This component, though less massive, dominates the optical cross section, and therefore appearance, of the disk."687" The fact that very large grains are so visible on a global scale in the Formalhaut disk means that small grains have been cleaned out from this disk extremely efficiently, and that all there is left are these very large grains, directly probed by reflected stellar light."," The fact that very large grains are so visible on a global scale in the Formalhaut disk means that small grains have been cleaned out from this disk extremely efficiently, and that all there is left are these very large grains, directly probed by reflected stellar light."688" We conclude that we are observing the transition part of parameter space, going from dust scattering to planetesimal reflection."," We conclude that we are observing the transition part of parameter space, going from dust scattering to planetesimal reflection."689" In fact, one might conclude that, almost 400 years after Galileo observed the crescentof Venus, we are seeing the crescents of the large dust grains inthe disk around Fomalhaut."," In fact, one might conclude that, almost 400 years after Galileo observed the crescentof Venus, we are seeing the crescents of the large dust grains inthe disk around Fomalhaut."690jn 2 OL. 1.5).,"bin z $\approx$ [0.1, 1.5]."691 Table 1. shows the various. properties of the subsamples alongwith their number densities., Table \ref{table1} shows the various properties of the subsamples alongwith their number densities.692 The galaxy number densities were computed. by integrating he luminosity functions. derived by Ibert οἱ al.," The galaxy number densities were computed by integrating the luminosity functions, derived by Ilbert et al."693 2005 on the same galaxy sample ane parameterized using Schechter functions., 2005 on the same galaxy sample and parameterized using Schechter functions.694 The evolution in the best-fit Schechter xwameters. M. ó* and a was taken into consideration. hereby accounting for luminosity evolution.," The evolution in the best-fit Schechter parameters, $M^*$, $\phi^*$ and $\alpha$ was taken into consideration, thereby accounting for luminosity evolution."695 We estimated he errors on the number densities by propagating the errors on the Schechter; parameters., We estimated the errors on the number densities by propagating the errors on the Schechter parameters.696" Alp""tores7"" denotes the evolving absolute magnitude threshold at the highest. redshift.", $M_B^{thresh}$ denotes the evolving absolute magnitude threshold at the highest redshift.697" For cach ALZ7"" two samples were obtained. one at low redshilt and another at higher recdshift with brighter galaxies selected due to the evolving selection eut."," For each $M_B^{thresh}$ two samples were obtained, one at low redshift and another at higher redshift with brighter galaxies selected due to the evolving selection cut."698 The two samples overlap slightlv in redshift in order to maximise the number of objects., The two samples overlap slightly in redshift in order to maximise the number of objects.699": ALB”lou and ALY""high are the absolute magnitudes. of the evolving eut at the lower and higher recshilt limits respectively of the redshift range.", $M_B^{low}$ and $M_B^{high}$ are the absolute magnitudes of the evolving cut at the lower and higher redshift limits respectively of the redshift range.700 Icleally. we would like to follow statistically the same galaxy population with time in order to study the growth of the underlving clark matter halo mass.," Ideally, we would like to follow statistically the same galaxy population with time in order to study the growth of the underlying dark matter halo mass."701 This is à. tricky issue as it is cillieult if not. impossible to know the exact progenitors of a descendant. galaxy population ancl how to select them., This is a tricky issue as it is difficult if not impossible to know the exact progenitors of a descendant galaxy population and how to select them.702 However. by taking care to follow themir down to a fixed. absolute. Iuminositv. we can minimize the bias in the average halo mass due to the presenceflack of faint or bright galaxies.," However, by taking care to follow the down to a fixed absolute luminosity, we can minimize the bias in the average halo mass due to the presence/lack of faint or bright galaxies."703 As mentioned above. this is mace possible by accurate measurements of ealaxy evolution.," As mentioned above, this is made possible by accurate measurements of galaxy evolution."704 In order to tackle this issue we use the Malli-Millennium simulation of galaxies having 270° particles in à box of 62.5Mpe/h on its side (Springeletal.2005)., In order to tackle this issue we use the Milli-Millennium simulation of galaxies having $270^3$ particles in a box of $62.5 Mpc/h$ on its side \citep{spr05}.705. The simulations retain information on the progenitor trees of galaxies making it possible to make a comparison with volume-Iimited samples., The simulations retain information on the progenitor trees of galaxies making it possible to make a comparison with volume-limited samples.706 Let us select a galaxy sample at high redshift. chosen with a luminosity eut-olf in the D-band. similar to what is done in the VVD$S data. hereafter called the parent sample’.," Let us select a galaxy sample at high redshift chosen with a luminosity cut-off in the B-band similar to what is done in the VVDS data, hereafter called the 'parent sample'."707" This sample is then evolved into two samples at ower redshift. a ""simulated sample! at the lower redshift. laving a luminosity cut-oll that is evolved and fainter (again similar to what we did in the data). and. another sample hat contains all the descendant. galaxies at the same lower redshift (hereafter the descendant sample)."," This sample is then evolved into two samples at lower redshift, a 'simulated sample' at the lower redshift, having a luminosity cut-off that is evolved and fainter (again similar to what we did in the data), and another sample that contains all the descendant galaxies at the same lower redshift (hereafter the 'descendant sample')."708 Doing a galaxy o galaxy match between the two samples would: tell us iow many galaxies in the simulated cata set are actua descendants and therefore the same population followec hrough time and the cllects on the underlving average ido mass., Doing a galaxy to galaxy match between the two samples would tell us how many galaxies in the simulated data set are actual descendants and therefore the same population followed through time and the effects on the underlying average halo mass.709 Table 2. shows the Millennium samples selecte aaving roughly the same mean redshift ancl mean absolute magnitude. Ade. as in the VWDS data sample of Table 1..," Table \ref{table2} shows the Millennium samples selected having roughly the same mean redshift and mean absolute magnitude, $M_B$, as in the VVDS data sample of Table \ref{table1}."710 We lind that at worst. ol the AL<20 and a rest ob the AL<18 simulated. luminosity thresholc sample are actual descendants at. lower redshifts., We find that at worst of the $M<-20$ and at best of the $M<-18$ simulated luminosity threshold sample are actual descendants at lower redshifts.711 From ‘Table 2. we can also study. the cllect of the selections on the underlying average halo mass., From Table \ref{table2} we can also study the effect of the selections on the underlying average halo mass.712 Ht can be seen tha typically the underlying halos in the descendant sample are heavier than those in the simulated sample., It can be seen that typically the underlying halos in the descendant sample are heavier than those in the simulated sample.713 After taking a closer look at the descendant. sample we noted that even though there are a larger number of fainter galaxies. there are also slightly more bright galaxies that lead to a slightly. higher average magnitude.," After taking a closer look at the descendant sample we noted that even though there are a larger number of fainter galaxies, there are also slightly more bright galaxies that lead to a slightly higher average magnitude."714 Phe combination of faint satellite ealaxies residing in massive halos and [fewer galaxies of intermediate Luminosity. likely lead to a descendant sample with more massive halos ancl slightly brighter galaxies on average than the simulated sample.," The combination of faint satellite galaxies residing in massive halos and fewer galaxies of intermediate luminosity, likely lead to a descendant sample with more massive halos and slightly brighter galaxies on average than the simulated sample."715 This possibly causes a lower overlap between the simulated ancl descendant samples [or the brightest samples., This possibly causes a lower overlap between the simulated and descendant samples for the brightest samples.716 The Millennium simulation shows that a growth in halo mass detected in the data would. be uncerestimatec with respect to what could be seen ideally., The Millennium simulation shows that a growth in halo mass detected in the data would be underestimated with respect to what could be seen ideally.717 The underestimation in mass is of the order of roughly LO %. and. therefore a measure in the growth of mass of a halo can be mainly attributed to the hierarchical formation of structure ancl not due to the typology of the selection (taking into consideration the high overlap between the simulated: aud descendant samples).," The underestimation in mass is of the order of roughly 10 $\%$, and therefore a measure in the growth of mass of a halo can be mainly attributed to the hierarchical formation of structure and not due to the typology of the selection (taking into consideration the high overlap between the simulated and descendant samples)."718" The redshift-space correlation functions for the dillerent luminosity threshold. samples have been computed via the Lancy Szalay (1993) estimator: where Ne and Nyy are respectively the total number of galaxies and randomly distributed. points in the same survey,", The redshift-space correlation functions for the different luminosity threshold samples have been computed via the Landy Szalay (1993) estimator: where $N_G$ and $N_R$ are respectively the total number of galaxies and randomly distributed points in the same survey.719" (ο(ντ) is the number of distinct. galaxy-galaxy pairs with scparations Wing in the interval (2.5 |dx) in the racial direction and Gry.r,| dr) perpendicular to the line of sight."," $GG(r_p,\pi)$ is the number of distinct galaxy-galaxy pairs with separations lying in the interval $\pi$ $\pi + d\pi$ ) in the radial direction and $r_p$ $r_p + d r_p$ ) perpendicular to the line of sight."720" Likewise. RAG.x) and ORO,7) are the number of rancomeorandom pairs ancl galaxy-rancdonm pairs respectively in the same interval."," Likewise, $RR(r_p,\pi)$ and $GR(r_p,\pi)$ are the number of random-random pairs and galaxy-random pairs respectively in the same interval."721 In order to avoid redshift space distortions. £(ry.3) has been integrated. along the line of sight to obtain the projected correlation function (Davis&Pechles1983): where is the real space correlation function. with LNmolar.," In order to avoid redshift space distortions, $\xi(r_p,\pi)$ has been integrated along the line of sight to obtain the projected correlation function \citep{dav83}: where $\xi(r)$ is the real space correlation function with $r$ = $\sqrt{r_p^2 + y^2}$."722 The measurements using the same sample impose a similar upper limit (Polloetal.2005.2006:Meneuxetal. 2006).," The measurements using the same sample impose a similar upper limit \citep{pol05, pol06, men06}."723. Pollo et al. (, Pollo et al. (724"2005) found that zy(r,) is quite insensitive tO ree in the range of 15Mpe/h«xus<25M1pefh For ey,i«LOM1pefh.",2005) found that $\omega_p(r_p)$ is quite insensitive to $\pi_{max}$ in the range of $15 Mpc/h < \pi_{max} < 25 Mpc/h$ for $r_p< 10 Mpc/h$.725 Too small a value for this limit would. cause an underestimation of the small-scale power. and too large a value would introduce noise.," Too small a value for this limit would cause an underestimation of the small-scale power, and too large a value would introduce noise."726" After several experiments. the optimal value of 24,220 sf has been acopted."," After several experiments, the optimal value of $\pi_{max}$ =20 $h$ has been adopted."727" The errors have been estimated. using bootstrap resampling of the data. which consists of computing the variance of w,(r,) in γω bootstrap realizations of the sample."," The errors have been estimated using bootstrap resampling of the data, which consists of computing the variance of $w_p(r_p)$ in $N_{real}$ bootstrap realizations of the sample."728 Each realization is obtained by randomly selecting a subset of galaxies from the data sample allowing for repetitions., Each realization is obtained by randomly selecting a subset of galaxies from the data sample allowing for repetitions.729 A correction factor is then applied to account for the underestimation of the errors obtained using this technique., A correction factor is then applied to account for the underestimation of the errors obtained using this technique.730 “Phis correction factor has been calibrated on mock samples to match the ensemble error (accounting for, This correction factor has been calibrated on mock samples to match the ensemble error (accounting for731determines clensity decrease in a region of |~2h...,determines density decrease in a region of $1 \sim 2 R_{\odot}$.732 The data ο CDS (SL2R.) and UVCS (~LGR.) exhibits the drastic decrease of density. which indicates (hat non-radial expansion is desired.," The data of CDS $\lesssim 1.2 733R_{\odot}$ ) and UVCS $\sim 1.6 R_{\odot}$ ) exhibits the drastic decrease of density, which indicates that non-radial expansion is desired."734 Adjustment of the other parameters of the flow tube geometry (rj and o: eq.(L4))) would eive the still better fit., Adjustment of the other parameters of the flow tube geometry $r_1$ and $\sigma$; \ref{eq:ftg}) )) would give the still better fit.735 Although we do not further search the best parameter set to lit to the observation. the figure indicates (hat our model could reproduce (he observed density profile bv the choice of the appropriate parameters (Fy.7(5—8)xIOerg em7s |! 7T£ 300s. ancl [μας& 5).," Although we do not further search the best parameter set to fit to the observation, the figure indicates that our model could reproduce the observed density profile by the choice of the appropriate parameters $F_{\rm w,0}\simeq (5-8)\times 10^5$ erg $^{-2}$ $^{-1}$, $\tau\approx 300$ s, and $f_{\rm max}\approx 5$ )."736 Fieure 9 compares our resulis of temperature distribution with observation in the streamer region., Figure \ref{fig:obste} compares our results of temperature distribution with observation in the streamer region.737 Our models employ (he same parameter sels as in fig.3.. and (he data were electron temperature obtained. [rom observationsof a line ratio of Fe NIII/Fe X by the CDS and UVCS/SOIO (Parentietal.2000).," Our models employ the same parameter sets as in \ref{fig:obsne}, and the data were electron temperature obtained from observationsof a line ratio of Fe XIII/Fe X by the CDS and UVCS/SOHO \citep{pbp00}."738. Although the observed data are electron temperature. (μον are supposed to represent the plasma temperature because electron-ion equilibrium is attained in dense streamer region of r<28. (Ravimondetal.1993).," Although the observed data are electron temperature, they are supposed to represent the plasma temperature because electron-ion equilibrium is attained in dense streamer region of $r\lesssim 2R_{\odot}$ \citep{rskn98}."739. Therefore. it is reasonable to compare them to the results of our model considering one-fluid coronal plasma.," Therefore, it is reasonable to compare them to the results of our model considering one-fluid coronal plasma."740 A case emploving (Fi.7.fina)=(7.8x10°.300.5) gives reasonable peak temperature of 1.5x 109. However. none of our models can reproduce the observed location of Z4.," A case employing $(F_{\rm w,0}, \tau, f_{\rm max})=(7.8\times 10^5, 300, 5)$ gives reasonable peak temperature of $\simeq 1.5 \times 10^6$ K. However, none of our models can reproduce the observed location of $T_{\rm max}$."741 While the location is observationally inferred to be between 1.2/2. and 1.6H... all of our models give rpg<124...," While the location is observationally inferred to be between $1.2 R_{\odot}$ and $1.6 R_{\odot}$, all of our models give $r_{T{\rm max}}< 1.2 R_{\odot}$."742 This is because the dissipation leneth of the is essentiallv short. even though one considers long period-waves (hat are generated in the corona.," This is because the dissipation length of the N-waves is essentially short, even though one considers long period-waves that are generated in the corona."743 We can summarize (hat acoustic waves excited in the corona could certainly heat the surrounding plasma to T> 1001. however. they cannot maintain the high temperature (ill the sufficiently. distant. region by themselves.," We can summarize that acoustic waves excited in the corona could certainly heat the surrounding plasma to $T>10^6$ K, however, they cannot maintain the high temperature till the sufficiently distant region by themselves."744 Therefore. the cooperations will other healing sources with larger dissipation length are necessary (o explain the observed. solar COLPOLnDa.," Therefore, the cooperations with other heating sources with larger dissipation length are necessary to explain the observed solar corona."745 In fig.10.. we show the results of velocity distribution of the solar wind. wilh observational results in the low-latitude streamer (shaded).," In \ref{fig:obsvl}, we show the results of velocity distribution of the solar wind, with observational results in the low-latitude streamer (shaded)."746 The observational data are [rom (1997).. who determined velocity profile between 2 and 30 2. from measurements of about 65 moving objects in the streamer belt.," The observational data are from \citet{she97}, who determined velocity profile between 2 and 30 $R_{\odot}$ from measurements of about 65 moving objects in the streamer belt."747 They used two different technic in deriving (he results. whereas (he shaded area displaved in fig.1Q are based on the straight-line fit method (an upper panel of fie.6 in Sheeleyetal. (1997): The shaded region is traced from that figure).," They used two different technic in deriving the results, whereas the shaded area displayed in \ref{fig:obsvl} are based on the straight-line fit method (an upper panel of fig.6 in \citet{she97}; ; The shaded region is traced from that figure)."748the mechanism of their formation is still open.,the mechanism of their formation is still open.749 Our understuxding of MIID processes on the Sun is getting significantly improved with the rapid progress in observational instruments. data analvsis. methods and numerical modeling.," Our understanding of MHD processes on the Sun is getting significantly improved with the rapid progress in observational instruments, data analysis, methods and numerical modeling."750 For example. the data obtained by helioseismology. have provided initial information about the structure ancl dynamics of convective [lows around sunspols and emerging magnetic flux beneath the solar surface (e.g.IXosovielievetal.2000:Zhaoetal.2001:Nosovichey 2009).," For example, the data obtained by helioseismology have provided initial information about the structure and dynamics of convective flows around sunspots and emerging magnetic flux beneath the solar surface \citep[e.g.][]{kosovichev2000,zhao2001,kosovichev09}."751. The high-resolution observations from ground-based telescopes and Ilinode space mission have provided detailed data about the filamentary magnetic structures and flow dvnamies on the surface (e.g.Ichimotoetal.2007:Bonetal.2008:Attieet2009:Dalmaceda 2010).," The high-resolution observations from ground-based telescopes and Hinode space mission have provided detailed data about the filamentary magnetic structures and flow dynamics on the surface \citep[e.g.][]{ichimoto07,bonet08,attie2009,balmaceda2010}."752. In addition. important support for the unclerstanding ancl interpretation of the observations is given by “realistic” radiative MUD numerical simulations. which are based on the first principles and take into account all essential physical processes.," In addition, important support for the understanding and interpretation of the observations is given by ""realistic"" radiative MHD numerical simulations, which are based on the first principles and take into account all essential physical processes."753 The recent progress in (he nunerical modeling has made 1 possible to reproduce in simulations many observational effects in the quiet Sun region. sunspols aud active regions (e.g.Stein&Nordlund2001:SehüsslerVoeler2006:Jacoutoletal.20082.b:Martínez-Svkora2008:NiGashvili2009a. 2010).. magnetic flux emerging (e.g.Shibataetal.1989:Cheung2008:Steinοἱ2009). and even the whole magnetic structures. such as pores and sunspots (e.g.Ixnólker&Schüssler.1983:Steinetal.2003:Bercik2002:Rempel 2009).," The recent progress in the numerical modeling has made it possible to reproduce in simulations many observational effects in the quiet Sun region, sunspots and active regions \citep[e.g.][]{stein2001,shussler2006,jacoutot08a,jacoutot08b,martinez08,kiti09a,kiti10}, magnetic flux emerging \citep[e.g.][]{shibata1989,cheung08,stein09} and even the whole magnetic structures, such as pores and sunspots \citep[e.g.][]{knolker1988,stein03,bercik2003,rempel2009}."754. However. most of the modeling has been done by settinge up the initial conditions with already existinge magnetice structures. e.g.e a horizontal fIux tube for the modeling of magnetic fIux emerging. or a vertical fIux tube with strong field for the sunspot/pore structures simulations.," However, most of the modeling has been done by setting up the initial conditions with already existing magnetic structures, e.g. a horizontal flux tube for the modeling of magnetic flux emerging, or a vertical flux tube with strong field for the sunspot/pore structures simulations."755 [t seems that so far only one study succeeded in reproducing a spontaneous formation of a micropore-like magnetic structure from an initially uniform field in the turbulent convection of the Sun (Steinetal.2003)., It seems that so far only one study succeeded in reproducing a spontaneous formation of a micropore-like magnetic structure from an initially uniform field in the turbulent convection of the Sun \citep{stein03}.756. llowever. the lifetime of this structure was rather short. only few convective turnover time scales” (Berciketal...2003).," However, the lifetime of this structure was rather short, only ""few convective turnover time scales"" \citep{bercik2003}."757. Similar calculations by Vogleretal...(2005) [or à substantially shallower convective laver did not show the structure formation., Similar calculations by \cite{vogler2005} for a substantially shallower convective layer did not show the structure formation.758 Ilere. we present new results of the realistic MIID simulations that show a process of spontaneous formation of a stable pore-like magnetic structure from an uniform magnetic [ield. and discuss the physical mechanism of the structure formation. ancl its dynamics and evolution.," Here, we present new results of the realistic MHD simulations that show a process of spontaneous formation of a stable pore-like magnetic structure from an uniform magnetic field, and discuss the physical mechanism of the structure formation, and its dynamics and evolution."759 For the simulations we used a 3D radiative MIID code. “SolarBox”. developed by A. Wrav al NASA Advanced Supercomputing Division," For the simulations we used a 3D radiative MHD code, ""SolarBox"", developed by A. Wray at NASA Advanced Supercomputing Division"760‘There have been Major recent advances in) cosmology with impact on galaxy formation theory.,There have been major recent advances in cosmology with impact on galaxy formation theory.761 These include the detection of the temperature.polarisation cross power spectrum for the CMD by WALAP (IXogut.ctal.2003).. and measurements both of the CAIB temperature. power spectrum and the underlving matter power spectrum with unprecedented: accuracy. utilizing the WNLAD. 2DE. and quasar Lye absorption line data sets (Bennettetal.2003:Spergelet.," These include the detection of the temperature–polarisation cross power spectrum for the CMB by WMAP \cite{kogut}, and measurements both of the CMB temperature power spectrum and the underlying matter power spectrum with unprecedented accuracy, utilizing the WMAP, 2DF and quasar $\alpha$ absorption line data sets \cite{bennett,spergel}."762al. 2003).. Two results that have received considerable attention are the optical depth of the universe To—0.1740.03. which requires that the epoch of reionisation occurs at z=15-20 from WALAP (lxogut.etal.2003: 2003)... and the rolling. spectral index dnfdlnk=0.03+0.01 for approximately scale-invariant density Ductuations For a combination of WNLAP. 2DE. and Lya data (Spergelctal.," Two results that have received considerable attention are the optical depth of the universe $\tau=0.17\pm 0.03,$ which requires that the epoch of reionisation occurs at z=15-20 from WMAP \cite{kogut,spergel}, and the rolling spectral index $\mathrm{d}n/\mathrm{d}lnk = -0.03 \pm 0.01$ for approximately scale-invariant density fluctuations for a combination of WMAP, 2DF and $\alpha$ data \cite{spergel}."7632003).. There ds some tension. between these results: if both are correct. dto ds dillicult/ to understand how recombination occurred so carly without some mocification of the canonical model of primordial. nearly. scale-cinvariant Gaussian adiabatic density Wuctuations (Ciardi..Ferrara&Lllolder2005:Somervilleοἱal. 2003).," There is some tension between these results: if both are correct, it is difficult to understand how recombination occurred so early without some modification of the canonical model of primordial, nearly scale-invariant Gaussian adiabatic density fluctuations \cite{ciardi,fukugita,haiman,som2}."764. In fact. a Dew Lya absorption data set [rom the SDSS C(qUASAES jw independently found evidence for a rolling spectral index (Seljak2003).. although an independent analysis of he same data does not reproduce sullicienthy small error xws to confirm this result (Abazajian.," In fact, a new $\alpha$ absorption data set from the SDSS quasars has independently found evidence for a rolling spectral index \cite{seljak}, although an independent analysis of the same data does not reproduce sufficiently small error bars to confirm this result \cite{abazajian}."765&Dodelson2003).. The Lye lines measure power in the underlving matter power spectrum. on a comoving scale of around. 1 Alpe., The $\alpha$ lines measure power in the underlying matter power spectrum on a comoving scale of around 1 Mpc.766 The results are however subject to. bias. since one has o be confident that the eas is relatively unperturbed! w feedback. such as is seen in the vicinity of Lyman wreak galaxies to Alpe distances.," The results are however subject to bias, since one has to be confident that the gas is relatively unperturbed by feedback, such as is seen in the vicinity of Lyman break galaxies to Mpc distances."767 Hence it is of particular interest to consider another measure of the power spectrum on even smaller comoving scales. 10 to 0.1 Alpe.," Hence it is of particular interest to consider another measure of the power spectrum on even smaller comoving scales, $ 10^{-2} $ to 0.1 Mpc."768 This comes from spectroscopic gravitational lensing of quasar emission line region on several scales. the magnification ratios requiring and constraining substructure in the massive lensing halos (Metealfοἱal.," This comes from spectroscopic gravitational lensing of quasar emission line region on several scales, the magnification ratios requiring and constraining substructure in the massive lensing halos \cite{met}."769"2003).. One needs substantial power in objects of 10"" to 10M... amounting to between 4 and 7 percent of the galaxy surface density. and this cannot easily be accommodated: in the usual CDM. models with standard. elliptical isothermal lens mass profiles."," One needs substantial power in objects of $10^6$ to $10^9 \rm M_\odot$, amounting to between 4 and 7 percent of the galaxy surface density, and this cannot easily be accommodated in the usual CDM models with standard elliptical isothermal lens mass profiles."770 Previous estimates of halo substructure from gravitational lensing using simple lens models are highly. uncertain (Dalal& 2002)., Previous estimates of halo substructure from gravitational lensing using simple lens models are highly uncertain \cite{dal}. .771 Moreover the numerical simulations, Moreover the numerical simulations772refsect:atoms..,\\ref{sect:atoms}.773 TiO and VO are the most Important opacity sources in M stars., TiO and VO are the most important opacity sources in M stars.774 They remain strong in the L dwarfs. but get weaker towards cooler temperature due to condensation into dust species like perovskite (CaTiOs). solid. titanium oxides and vanadium oxides. respectivelyV.," They remain strong in the L dwarfs, but get weaker towards cooler temperature due to condensation into dust species like perovskite $_3$ ), solid titanium oxides and vanadium oxides, respectively."775"jwfOpacOS5. Our spectra cover the TIO bandheads atAA..AA..ΑΑ., AA.. and AA."," Our spectra cover the TiO bandheads at, and ."776. All bands are recovered by the models. but the TiO bands are too strong in the theoretical spectra particularly in the L2 and LS dwarfs.," All bands are recovered by the models, but the TiO bands are too strong in the theoretical spectra particularly in the L2 and L5 dwarfs."777 This hints at remaining shortcomings m our treatment of dust settling (see refsect:models)). which evidently predicts too much TIO to be left 1n the atmosphere at temperatures between 1500 and KK. VO bands are visible atAA..AA.. and AA.," This hints at remaining shortcomings in our treatment of dust settling (see \\ref{sect:models}) ), which evidently predicts too much TiO to be left in the atmosphere at temperatures between 1500 and K. VO bands are visible at, and ."778. In general. VO bands match the data quite well. but a detailed comparison is hampered buy the mismatch of the TiO bands. which often affects the same spectral regions.," In general, VO bands match the data quite well, but a detailed comparison is hampered buy the mismatch of the TiO bands, which often affects the same spectral regions."779 Another uncertainty that may be responsible for parts of the mismatch between our data and the model spectra are the (TIO) absorption bands” oscillator strengths2005)., Another uncertainty that may be responsible for parts of the mismatch between our data and the model spectra are the (TiO) absorption bands' oscillator strengths.780. Laboratory experiments do not yet provide values at the accuracy required in order to match the data. and further improvement on the molecular data is needed.," Laboratory experiments do not yet provide values at the accuracy required in order to match the data, and further improvement on the molecular data is needed."781 The three metal hydride species CaH. FeH. and CrH are important in the optical spectra of L dwarfs.," The three metal hydride species CaH, FeH, and CrH are important in the optical spectra of L dwarfs."782 CaH at us visible in the L dwarfs and is accurately reproduced by the models., CaH at is visible in the L dwarfs and is accurately reproduced by the models.783 The excellent match between the LO model and the spectrum shows that the high frequency patterns visible at this spectral resolution (around AA)) 1s the structure of the absorption band and not noise., The excellent match between the L0 model and the spectrum shows that the high frequency patterns visible at this spectral resolution (around ) is the structure of the absorption band and not noise.784 These patterns disappear around LS. confirming that CaH is depleted at later spectral types and 1s superseded by the satellite feature (see refsect:alkalis)).," These patterns disappear around L5, confirming that CaH is depleted at later spectral types and is superseded by the satellite feature (see \\ref{sect:alkalis}) )."785 The general structure of the CrH band at aalso matches quite well the structure of the data in all three L dwarfs and also in the T dwarf., The general structure of the CrH band at also matches quite well the structure of the data in all three L dwarfs and also in the T dwarf.786 Unfortunately. this band is embedded in the strong TIO and VO bands. which the models overpredict. as well as FeH. The calculated strength of the CrH hence is difficult to judge in the L dwarfs (see below).," Unfortunately, this band is embedded in the strong TiO and VO bands, which the models overpredict, as well as FeH. The calculated strength of the CrH hence is difficult to judge in the L dwarfs (see below)."787 FeH ts observed close to the CrH band at aand extends to the red all the way to >μπα. but is increasingly superposed by CrH. and between bby TiO. The models compare to the data as good as for CrH. The prominent Wing-Ford band of FeH is visible atAA.," FeH is observed close to the CrH band at and extends to the red all the way to $>1\mu$ m, but is increasingly superposed by CrH, and between by TiO. The models compare to the data as good as for CrH. The prominent Wing-Ford band of FeH is visible at."788. Structure and intensity of this band are well fit by the model at the first fewAngstrom. in early L dwarfs.," Structure and intensity of this band are well fit by the model at the first few, in early L dwarfs."789 However. this band is overestimated at very low temperature as well.," However, this band is overestimated at very low temperature as well."790 AtAA.. another CrH band should be visible.," At, another CrH band should be visible."791 It is contained in the model spectra dominating the absorption redwards of iin the L5 and TI spectra., It is contained in the model spectra dominating the absorption redwards of in the L5 and T1 spectra.792 However. show that all absorption features at that wavelength region can be explained by FeH in M-dwarfs. they see no evidence for CrH absorption.," However, show that all absorption features at that wavelength region can be explained by FeH in M-dwarfs, they see no evidence for CrH absorption."793 In the models shown here. CrH produces a strong step 1n the model spectra for the L5 and TI objects.," In the models shown here, CrH produces a strong step in the model spectra for the L5 and T1 objects."794 Such a feature at iis not observed in our data. or at least at much less strength. supporting the argument that the absorption around jm in early L dwarfs is entirely due to FeH. and that CrH is much weaker than predicted.," Such a feature at is not observed in our data, or at least at much less strength, supporting the argument that the absorption around $\mu$ m in early L dwarfs is entirely due to FeH, and that CrH is much weaker than predicted."795 A similar case can be made for the bband., A similar case can be made for the band.796 While for the earlier Ls the agreement with the observed band seems reasonably good. in the T dwarf system it is much too strong in the model.," While for the earlier Ls the agreement with the observed band seems reasonably good, in the T dwarf system it is much too strong in the model."797 A more detailed investigation of the molecular line formation in the models reveals that at higher 7; both FeH and CrH contribute in similar shares to the absorption all the way to the blue end. with FeH being mostly responsible for the stronger lines and the structure of the band.," A more detailed investigation of the molecular line formation in the models reveals that at higher $T_{\rm eff}$ both FeH and CrH contribute in similar shares to the absorption all the way to the blue end, with FeH being mostly responsible for the stronger lines and the structure of the band."798 In the KK model however. settling effects have already removed significant quantities of the hydrides from the line forming region.," In the K model however, settling effects have already removed significant quantities of the hydrides from the line forming region."799 FeH ts much stronger affected by this. beingορ depleted already two pressure scale heights deeper than CrH. which is equivalent to about one order of magnitude in optical depth.," FeH is much stronger affected by this, being depleted already two pressure scale heights deeper than CrH, which is equivalent to about one order of magnitude in optical depth."800 Thus. in the ΤΙ dwarf the band ts probably dominated by CrH. We therefore conclude that we see the combined effects of FeH with fairly accurate opacity data. and CrH absorption that appears overestimated by a significant factor. in our spectral sequence.," Thus, in the T1 dwarf the band is probably dominated by CrH. We therefore conclude that we see the combined effects of FeH with fairly accurate opacity data, and CrH absorption that appears overestimated by a significant factor, in our spectral sequence."801 Future modelling efforts for improvement in this spectral region should therefore focus on better oscillator strengths for the CrH lines., Future modelling efforts for improvement in this spectral region should therefore focus on better oscillator strengths for the CrH lines.802 Strong steam bands are visible at very low temperatures in the wavelength region redward ofAA.. where TiO and CrH are also important (see above).," Strong steam bands are visible at very low temperatures in the wavelength region redward of, where TiO and CrH are also important (see above)."803 The, The8041 fractional polarization insteac of the drop. observed ‘lose to the radio outburst.,the fractional polarization instead of the drop observed close to the radio outburst.805 Llowever. we must note that if the shock is oblique instead. of transverse. the expected variations in the polarization properties are clillerent. and strongly related to the obliqueness of the shock itself and to the characteristics of the underling magnetic field. like its order and strength.," However, we must note that if the shock is oblique instead of transverse, the expected variations in the polarization properties are different, and strongly related to the obliqueness of the shock itself, and to the characteristics of the underling magnetic field like its order and strength."806 Therefore. a reliable interpretation of the observed. polarization trends requires a much better sampling than that available between 1998-2001. (Fig. 6)).," Therefore, a reliable interpretation of the observed polarization trends requires a much better sampling than that available between 1998-2001 (Fig. \ref{plot_mojave99}) ),"807 leaving the debate on the nature of the main mechanism at work still The analysis of the lightcurves ancl polarization trends between 1998 ancl 2001. shown in Fig. 6..," leaving the debate on the nature of the main mechanism at work still The analysis of the lightcurves and polarization trends between 1998 and 2001, shown in Fig. \ref{plot_mojave99},"808 suggests that the variations of the total intensity. [lux density and the polarization properties may be explained. by both the previous scenarios., suggests that the variations of the total intensity flux density and the polarization properties may be explained by both the previous scenarios.809 Indeed. the total intensity and polarization properties can be related to the evolution (likely aciabatic expansion) of either a new jet component or a shock originated at the beginning of 1999.," Indeed, the total intensity and polarization properties can be related to the evolution (likely adiabatic expansion) of either a new jet component or a shock originated at the beginning of 1999."810 Support to this interpretation comes from the detection of a superluminal component that possibly originated in. 1999.1040.25 (Listeretal. 2009b)., Support to this interpretation comes from the detection of a superluminal component that possibly originated in $\pm$ 0.25 \citep{lister09b}.811. The rotation of the angle Found in 2001 may be explained. considering that this new component moves far enough from the core to be resolved and the intrinsic polarized emission of the core can be separated. [rom that of the The lighteurves considered in the above explanation would require a more frequent sampling and for this reason a definitive Interpretation of the physical mechanisms at work is precluced., The rotation of the angle found in 2001 may be explained considering that this new component moves far enough from the core to be resolved and the intrinsic polarized emission of the core can be separated from that of the The lightcurves considered in the above explanation would require a more frequent sampling and for this reason a definitive interpretation of the physical mechanisms at work is precluded.812 A similar consideration may apply to explain the source behaviour around. April 2009. when information at other wavelengths. (c.g.Abdoetal.2010b:Marscheretal.2010). is available.," A similar consideration may apply to explain the source behaviour around April 2009, when information at other wavelengths \citep[e.g.][]{abdo10b,marscher10} is available."813 The multiwavelength lightcurve presented by Abdoctal.(2010b) shows a flux density enhancement around that. period that is first detected a 230 Gllz. and after some delay. at lower frequencies.," The multiwavelength lightcurve presented by \citet{abdo10b} shows a flux density enhancement around that period that is first detected at 230 GHz, and after some delay at lower frequencies."814 The same behaviour is visible in Fie., The same behaviour is visible in Fig.815 Ll where the lighteurves at 15 and 43 have been compared., \ref{peak_flux} where the lightcurves at 15 and 43 have been compared.816 From Fig., From Fig.817 1 we see that the Hux densities at. both frequencies. have a similar behaviour. where the 43-CGllz data points seem to anticipate those at. lower frequency.," \ref{peak_flux} we see that the flux densities at both frequencies have a similar behaviour, where the 43-GHz data points seem to anticipate those at lower frequency."818 Interestingly. jus after. March. 2009. when several 5-rav Lares have been (etectecd (DAmimancoetal.2011).. the tux density a 43 Gllz becomes higher than that at 15 CGllz implying a change in the opacitv of the component.," Interestingly, just after March 2009, when several $\gamma$ -ray flares have been detected \citep{dammando11}, the flux density at 43 GHz becomes higher than that at 15 GHz implying a change in the opacity of the component."819 The maximum of the radio emission is then reached. in April 2009 when another strong 5-rav [lare was detected (Cutini&Llavs 2009)., The maximum of the radio emission is then reached in April 2009 when another strong $\gamma$ -ray flare was detected \citep{cutini09}.820. In. correspondence of this luminosity enhancement 16 polarization percentage drops while the EWPA changes by about 75 (Fig. 10)), In correspondence of this luminosity enhancement the polarization percentage drops while the EVPA changes by about $^{\circ}$ (Fig. \ref{multi_chi}) )821 at both 15 and 43 Gllz becoming parallel to the jet. direction., at both 15 and 43 GHz becoming parallel to the jet direction.822 In the same period a strong rotation of the optical polarization vector (Alarscherctal.2010) has been detected. together with the ejection of a new superluminal component observed. at 15 Cllz (Fig. S)).," In the same period a strong rotation of the optical polarization vector \citep{marscher10} has been detected together with the ejection of a new superluminal component observed at 15 GHz (Fig. \ref{fit_mojave}) ),"823 and already reported by Abdoctal.(2010b).. and. at 43 CGllz (Marscherctal.2010).," and already reported by \citet{abdo10b}, and at 43 GHz \citep{marscher10}."824. A similar example of dramatic rotation of the polarization angle and a drop of the fractional polarization in coincidence with a 5-ray fare was found in the blazar 2279. indicating a co-spatiality of the optical and *-rav emission region (Abdoetal.20100).," A similar example of dramatic rotation of the polarization angle and a drop of the fractional polarization in coincidence with a $\gamma$ -ray flare was found in the blazar 279, indicating a co-spatiality of the optical and $\gamma$ -ray emission region \citep{abdo10c}."825".. ""Ehis has been explained assuming a non-axisvmametrie structure of the emitting region. Likely a curved jet trajectory. rather than a perpendicular shock/ moving in an axially svnunetric jet as we suggested in Section 4.1."," This has been explained assuming a non-axisymmetric structure of the emitting region, likely a curved jet trajectory, rather than a perpendicular shock moving in an axially symmetric jet as we suggested in Section 4.1."826 In this case. he degree and angle of polarization strictly depends on the instantaneous angle formed. by the direction of the motion with our line of sight (Abdoetal.2010c).," In this case, the degree and angle of polarization strictly depends on the instantaneous angle formed by the direction of the motion with our line of sight \citep{abdo10c}."827. The similarity oween. 302279 and suggests à. common origin also for the emission at11510-0809. high energy and in the racio xuxd. and the time [ag in the Lux density behaviour may be due to opacity ellects as the shock passes by.," The similarity between 279 and 1510-089 suggests a common origin also for the emission at high energy and in the radio band, and the time lag in the flux density behaviour may be due to opacity effects as the shock passes by."828 A cospatiality of the 5-rav and radio emitting region was also claimed or the BL Lac object 2287. where two ταν [laring episodes occurred. close in time with two major millimeter outbursts (Xgudoetal.2011).," A cospatiality of the $\gamma$ -ray and radio emitting region was also claimed for the BL Lac object 287, where two $\gamma$ -ray flaring episodes occurred close in time with two major millimeter outbursts \citep{agudo11}."829. In 11510-089. the variation in the polarization angle at both 15 and 43 Gllz suggests a change in the magnetic field orientation in the compact component rather than Faraday ellects caused by an external screen.," In 1510-089, the variation in the polarization angle at both 15 and 43 GHz suggests a change in the magnetic field orientation in the compact component rather than Faraday effects caused by an external screen."830 Changes in the polarization angle of the same magnitude at dillerent wavelengths were reported. by Homanctal.(2002h) who monitored the behaviour of a sample of 12 blazars bv means of dual-Frequeney. VLBA ]t is worth noting that. as in the case of 302279 where no changes in the racio band could be found related to the 5- [lare (Abcloctal.2010€).. in 11510-089 no obvious connection between the 5-rav activity detected in. March 2009 (D'Ammandoctal.2011). and the radio lux density behaviour has been found. suggesting that during this [are the svnchrotron radiation in the radio band is not vet fully optically thin.," Changes in the polarization angle of the same magnitude at different wavelengths were reported by \citet{homan02a} who monitored the behaviour of a sample of 12 blazars by means of dual-frequency VLBA It is worth noting that, as in the case of 279 where no changes in the radio band could be found related to the $\gamma$ -ray flare \citep{abdo10c}, in 1510-089 no obvious connection between the $\gamma$ -ray activity detected in March 2009 \citep{dammando11} and the radio flux density behaviour has been found, suggesting that during this flare the synchrotron radiation in the radio band is not yet fully optically thin."831 We have presented results from the analysis of multi-epoch polarimetric VLBI. Space-VLBIE and archival VLBA data from the MOJAVE programme of the flat spectrum racio quasar 11510-089 spanning over 15 vears (1995-2010).," We have presented results from the analysis of multi-epoch polarimetric VLBI, Space-VLBI and archival VLBA data from the MOJAVE programme of the flat spectrum radio quasar 1510-089 spanning over 15 years (1995-2010)."832 This source shows a pe-scale core-jet structure where superluminal knots are ejectecl at. cilferent times., This source shows a pc-scale core-jet structure where superluminal knots are ejected at different times.833 From the multi-epoch observations we found that. the emission of new blobs is roughly constant with a time lag of about one vear., From the multi-epoch observations we found that the emission of new blobs is roughly constant with a time lag of about one year.834 Furthermore. the various jet components. are moving away from the core with an apparent superluminal speed in the range between 15e and. 20e and roughly with the same position angle. suggesting that the precession of the jet is not. relevant on the timescale of decades. in our frame.," Furthermore, the various jet components are moving away from the core with an apparent superluminal speed in the range between $c$ and $c$ and roughly with the same position angle, suggesting that the precession of the jet is not relevant on the timescale of decades in our frame."835 Both the total intensity and. the polarized Hux censity of the core component show high level of variability., Both the total intensity and the polarized flux density of the core component show high level of variability.836 Our analysis shows that occasionally the EVPA, Our analysis shows that occasionally the EVPA837(1996).,(1996).838 Five multislit masks were prepared. using the MOS software provided at the CELUE. though only four were actually used.," Five multislit masks were prepared using the MOS software provided at the CFHT, though only four were actually used."839 Slits were cut using the machine. with typically 40 ~ 45 objects per mask.," Slits were cut using the machine, with typically 40 $\sim$ 45 objects per mask."840 One of the masks was centred. on the nucleus of NGC 4472. with the other four being displaced by ~ 5:55 into the NE. NW. SE. and SW quacdrants to obtain the best spatial coverage of the cluster system.," One of the masks was centred on the nucleus of NGC 4472, with the other four being displaced by $\sim$ 5 into the NE, NW, SE, and SW quadrants to obtain the best spatial coverage of the cluster system."841 Llighest priority was given to candidates in the magnitude range 19.5«V21.5. although candidates were selected. down ," Highest priority was given to candidates in the magnitude range $19.5< V <21.5$, although candidates were selected down to $V$ =22.5."842‘Table 1. gives the details of the masks. listing the number of candidates per mask. alone with the number of spectroscopically confirmed elobular clusters. background galaxies and foreground stars.," Table \ref{tab:masks}843 gives the details of the masks, listing the number of candidates per mask, along with the number of spectroscopically confirmed globular clusters, background galaxies and foreground stars."844 A number of spectra vielded no reliable identification. due to insullicient signal-to-noise (S/N) and/or problematical skv-subtraction.," A number of spectra yielded no reliable identification, due to insufficient signal-to-noise (S/N) and/or problematical sky-subtraction."845 The D600 erism with dispersion 2.24 | was used. producing spectra with an instrumental resolution of  5.5 (330 1) and a useful spectral range of 3800.— 6500.," The B600 grism with dispersion 2.24 $^{-1}$ was used, producing spectra with an instrumental resolution of $\sim$ 5.5 (330 $^{-1}$ ) and a useful spectral range of 3800 – 6500."846Α.. The detector was a STIS 2048? chip. with reacout noise 9.3 . noteworthy for its excellent. performance in the blue (quantum ellicieney ~ 8S2 at 4000 AY).," The detector was a STIS $^{2}$ chip, with readout noise 9.3 $^{-}$, noteworthy for its excellent performance in the blue (quantum efficiency $\sim$ 82 at 4000 )."847 Flat field and bias rames were taken at the beginning and end of cach night. and the spectra were wavelength. calibrated. using frequent mercury ares taken before and after the programme object Tames.," Flat field and bias frames were taken at the beginning and end of each night, and the spectra were wavelength calibrated using frequent mercury arcs taken before and after the programme object frames."848 Table 2. lists the observational details for the CLIT run., Table \ref{tab:obslog} lists the observational details for the CFHT run.849 For velocity. calibration. we have obtained long slit spectra for a number of radial velocity. standard: stars.," For velocity calibration, we have obtained long slit spectra for a number of radial velocity standard stars."850 In addition. we have taken integrated: spectra of several Galactic Globular Clusters (GC's) in order to calibrate metallicities.," In addition, we have taken integrated spectra of several Galactic Globular Clusters (GGCs) in order to calibrate metallicities."851 The spatial extent of these. globular clusters on the sky. (mean core radii  40 aresec) required. us to synthesize an aperture in order to obtain a representative integrated spectrum., The spatial extent of these globular clusters on the sky (mean core radii $\sim$ 40 arcsec) required us to synthesize an aperture in order to obtain a representative integrated spectrum.852 We have therefore scanned. the cores of the globular clusters over a typical range of 90 arcsec., We have therefore scanned the cores of the globular clusters over a typical range of 90 arcsec.853 Combined. with the WIIT observations. this vields high S/N spectra (~ 500 at 5000 9). of five GGCs in. the metallicity range -2.24 < Fe/H] < -0.29 dex with one overlap. namely NGC 6356.," Combined with the WHT observations, this yields high S/N spectra $\sim$ 500 at 5000 ) of five GGCs in the metallicity range -2.24 $<$ [Fe/H] $<$ -0.29 dex with one overlap, namely NGC 6356."854 We summarize our calibration objects in Table 3.., We summarize our calibration objects in Table \ref{tab:standards}.855 The majority of the data reduction was performed using the package in together with other standard tasks., The majority of the data reduction was performed using the package in together with other standard tasks.856 Object frames were trimmed. ancl biasesubtracted. and any bad. pixels were cleaned by interpolating across adjacent columns.," Object frames were trimmed and bias-subtracted, and any bad pixels were cleaned by interpolating across adjacent columns."857 A second-order polynomial fit to these data produced. residuals of order  0.1 iin the wavelength calibration., A second-order polynomial fit to these data produced residuals of order $\sim$ 0.1 in the wavelength calibration.858 Phe spectra were. then optimally extracted and sky-subtracted using a linear least-squares fit to the background sky., The spectra were then optimally extracted and sky-subtracted using a linear least-squares fit to the background sky.859 Finally. the spectra were rebinned with a step size of 2 oon to a logarithmic wavelength scale over the range 3800 5500A.," Finally, the spectra were rebinned with a step size of 2 on to a logarithmic wavelength scale over the range 3800 – 5500."860 The resultant. spectra typically possessed. S/N ratios of 4 10., The resultant spectra typically possessed S/N ratios of 4 – 10.861 The Fourier cross-correlation task was used. for determining racial velocities from the spectra., The Fourier cross-correlation task was used for determining radial velocities from the spectra.862 “Phe six velocity. templates were ce-recdshiftecl by their literature values and then individually cross-correlatecd against each of the candidate cluster spectra., The six velocity templates were de-redshifted by their literature values and then individually cross-correlated against each of the candidate cluster spectra.863 An rr value of 2.5 (Tonry Davis 1979) was set as the lower threshold for reliable measurement: below this any velocities returned. however plausible. were removed. from. further analysis.," An 'r' value of 2.5 (Tonry Davis 1979) was set as the lower threshold for reliable measurement; below this any velocities returned, however plausible, were removed from further analysis."864 The final velocities of the globular clusters were taken to be the mean velocity weighted by the eross-correlation. peak height. of each template., The final velocities of the globular clusters were taken to be the mean velocity weighted by the cross-correlation peak height of each template.865 A heliocentric correction was then applied to these velocities., A heliocentric correction was then applied to these velocities.866 In addition to the formal errors returned by the cross-correlation task. an estimate of the uncertainties mav be made from the overlap between common objects in dilferent masks.," In addition to the formal errors returned by the cross-correlation task, an estimate of the uncertainties may be made from the overlap between common objects in different masks."867 Between the data from Sharples (1998) and the CELT velocities there are à total of 13 overlaps., Between the data from Sharples (1998) and the CFHT velocities there are a total of 13 overlaps.868 The mean dillerence of the sample is -17 | with an inferred velocity uncertainty for a single measurement of TS , The mean difference of the sample is -17 $^{-1}$ with an inferred velocity uncertainty for a single measurement of 78 $^{-1}$.869We take a velocity range 300 Vix 2000 kms as being representative of globular clusters associated. witFoe NGC 4472. which has a heliocentric velocity of 961 kms (Sandage “Tammann 1981).," We take a velocity range 300 $\le V_{\rm{h}} \le$ 2000 $^{-1}$ as being representative of globular clusters associated with NGC 4472, which has a heliocentric velocity of 961 $^{-1}$ (Sandage Tammann 1981)."870 This is consistent with the mean velocity of our elobular cluster sample of 990 x 26 which has a velocity dispersion. of 314 ," This is consistent with the mean velocity of our globular cluster sample of 990 $\pm$ 26 $^{-1}$, which has a velocity dispersion of 314 $^{-1}$."871See Zepl (in preparation) For further analysis of the elobular cluster kinematics., See Zepf (in preparation) for further analysis of the globular cluster kinematics.872 Whilst the quality of our individual spectra is adequate to obtain radial velocities. it ds insullicicnt for. reliable line-streneth analysis.," Whilst the quality of our individual spectra is adequate to obtain radial velocities, it is insufficient for reliable line-strength analysis."873 For any believable. measurement of equivalent widths of absorption lines. a method of co-addition of the spectra was required so as to improve the S/N ratio.," For any believable measurement of equivalent widths of absorption lines, a method of co-addition of the spectra was required so as to improve the S/N ratio."874 Since the broadband: colours of the globular clusters »imarilv reflect their metallicitics (on the assumption that hese are old stellar populations. 7z 8 Car). blue globular clusters should. be metal-poor. and become progressively more metal-rich as they redden.," Since the broadband colours of the globular clusters primarily reflect their metallicities (on the assumption that these are old stellar populations, $\tau \ga$ 8 Gyr), blue globular clusters should be metal-poor, and become progressively more metal-rich as they redden."875 Phe cluster spectra were herefore assigned bins on the basis of their €*—T4 colours rom the photometry of Geisler (1996)., The cluster spectra were therefore assigned bins on the basis of their $C-T_{1}$ colours from the photometry of Geisler (1996).876 Phe globular cluster catalogue of Geisler (1996) is bimodal in colour. and the subset of brightest globular clusters selected rom this catalogue were chosen so as to rellect this umocdality. albeit with reduced. cluster numbers.," The globular cluster catalogue of Geisler (1996) is bimodal in colour, and the subset of brightest globular clusters selected from this catalogue were chosen so as to reflect this bimodality, albeit with reduced cluster numbers."877 Fig., Fig.878 1 shows the sample of S60. cluster candidates from Ceisler (1996) (open histogram) and. those for which racial velocities have been obtained and subsequently used in the metallicity analysis (shaded. histogram)., \ref{fig:bimodal} shows the sample of 860 cluster candidates from Geisler (1996) (open histogram) and those for which radial velocities have been obtained and subsequently used in the metallicity analysis (shaded histogram).879 Although we now have velocities for 141 globular clusters. 10 of these were from the original sample of Mould. (1990) and these spectra were unavailable for further analysis.," Although we now have velocities for 141 globular clusters, 10 of these were from the original sample of Mould (1990) and these spectra were unavailable for further analysis."880 Due to this bimodality in these data. creating bins of a," Due to this bimodality in these data, creating bins of a"881dise f: a) an inner part where electron scattering determines the opacity and radiation pressure is larger than eas pressure: b) a middle region where electron scattering is still more mnüportanut than free-free absorption. but gas pressure dominates over radiation pressure: ο) an outer region. where thermal eas pressure donates and free-free absorption is the main source of opacity. ,"disc \cite{ss73}: a) an inner part where electron scattering determines the opacity and radiation pressure is larger than gas pressure; b) a middle region where electron scattering is still more important than free-free absorption, but gas pressure dominates over radiation pressure; c) an outer region, where thermal gas pressure dominates and free-free absorption is the main source of opacity. }"882At preseut magneto-rotational instability (MIRI! is favoured as the primary source of the turbulent viscosity needed to explain the luuinositics of accreting black holes., At present magneto-rotational instability \cite{bh98} is favoured as the primary source of the turbulent viscosity needed to explain the luminosities of accreting black holes.883 Our knowledge of the plivsics of pdMIID turbulence in accretion discs can be used to build a selt-consisteut model of MRdriven. thin accretion svstenis iu the following way.," Our knowledge of the physics of MHD turbulence in accretion discs can be used to build a self-consistent model of MRI-driven, thin accretion disc--corona systems in the following way."884 Let us asstune equipartition between kinetic turbulent οσον aud magnetic field excited by the MRI (Alfvéónu equal to turbulent speed. ον=0).," Let us assume equipartition between kinetic turbulent energy and magnetic field excited by the MRI (Alfvénn equal to turbulent speed, $v_{\rm885A}=v_{\rm t}$ )."886" For alinost iucompressible flows+? the coefficieut of turbulent viscosity »702/30. which implies Q,=Septcas."," For almost incompressible flows \cite{pbb02} the coefficient of turbulent viscosity $\nu \simeq v_{\rm t}^2/3\Omega$, which implies $Q_+=\frac{3}{2}c_{\rm887 s} \nu \rho \Omega=c_{\rm s} P_{\rm mag}$."888 Therefore. ounce the relationship between magnetic and disce pressure (giveu either by eas or radiation) is established. the accretion disc structure can be fully described.," Therefore, once the relationship between magnetic and disc pressure (given either by gas or radiation) is established, the accretion disc structure can be fully described."889" To find such a relationship. we can assume *? that the magnetic field escapes from the thin dise via buovaney. with a timescale fj,—IT/204."," To find such a relationship, we can assume \cite{me02} that the magnetic field escapes from the thin disc via buoyancy, with a timescale $t_{\rm890 b}=H/2v_{\rm A}$."891" Then. a crucial point is that the growth rate of AIRT is influenced by the ratio of the eas to magnetic pressure 12.. a= Οσον. where e, is the gas sound speed."," Then, a crucial point is that the growth rate of MRI is influenced by the ratio of the gas to magnetic pressure \cite{bs01}: $\sigma=\Omega c_{\rm892g}/v_{\rm A}$ , where $c_{\rm g}$ is the gas sound speed."893" The saturation field can be found by noting that. asviuptotically. ofi,=O(1): this automatically gives the desired scaling for the magnetic pressure in MBI donunated turbulent flows: where ay=(+4? is a coustaut. not necessarily aller thau wnity."," The saturation field can be found by noting that, asymptotically, $\sigma t_{\rm894b}=O(1)$; this automatically gives the desired scaling for the magnetic pressure in MRI dominated turbulent flows: where $\alpha_0=(\frac{P_{\rm tot}}{\beta^2 P_{\rm gas}})^{1/2}$ is a constant, not necessarily smaller than unity."895 The magnetic flux escapingim the vertical direction may dissipate a substautial fraction of the eravitational binding οσον of the accreting easoutsidethe optically thick disc. with obvious deep müplicatious for the spectrum of the eniereius radiation.," The magnetic flux escaping in the vertical direction may dissipate a substantial fraction of the gravitational binding energy of the accreting gasthe optically thick disc, with obvious deep implications for the spectrum of the emerging radiation."896" The fraction f of the total power dissipated iu the low- envirounieut above aud below the disc (in the so-called corona) is determined by the ratio of the vertical Povuting flux (Fp=e4 Pua) to the local heating rate Q,."," The fraction $f$ of the total power dissipated in the low-density environment above and below the disc (in the so-called ) is determined by the ratio of the vertical Poynting flux $F_{\rm897P}\simeq v_{\rm A}P_{\rm mag}$ ) to the local heating rate $Q_+$ ."898 Cuder the assmuption of equipartition between turbuleut aud magnetic energies. this translates mto," Under the assumption of equipartition between turbulent and magnetic energies, this translates into"899near/far determinations for all assigned kinematic distances inverted.,near/far determinations for all assigned kinematic distances inverted.900 This is undertaken to examine the effect of incorrect near/far distance ambiguity assignments on the luminosity distributions., This is undertaken to examine the effect of incorrect near/far distance ambiguity assignments on the luminosity distributions.901 As can be see. this does not change the overall shape of the luminosity distributions appreciably.," As can be see, this does not change the overall shape of the luminosity distributions appreciably."902 Therefore the remaining individual uncertainties in the distances to the RMS sample do not significantly affect the global results on a statistical level., Therefore the remaining individual uncertainties in the distances to the RMS sample do not significantly affect the global results on a statistical level.903 These results will be discussed in detail in refS:discussion.., These results will be discussed in detail in \\ref{S:discussion}.904 Though far-IR fluxes were obtained by Mottrametal.(2010) for the majority of young RMS sources. this was not possible for some sources due to confusion within the available data. which in turn prevented SED fitting of these sources (see refS:testsyata)).," Though far-IR fluxes were obtained by \citet[][]{Mottram2010} for the majority of young RMS sources, this was not possible for some sources due to confusion within the available data, which in turn prevented SED fitting of these sources (see \\ref{S:tests_data}) )."905 However. thepropertiesofthosesourceswhere fitswereobtainedcanbeusedtocalculatethemeanratiobetweenthebolometric f flux for all sources with good SED fits.," However, the properties of those sources where fits were obtained can be used to calculate the mean ratio between the bolometric flux and the MSX flux for all sources with good SED fits."906 This mean ratio cai then be used to obtain estimates of the bolometric flux of RMS sources which do not have far-IR data. since all have measured MSX fluxes.," This mean ratio can then be used to obtain estimates of the bolometric flux of RMS sources which do not have far-IR data, since all have measured MSX fluxes."907 In addition to the total flux. filter-band fluxes were output for the SED model fits to the data.," In addition to the total flux, filter-band fluxes were output for the SED model fits to the data."908 The MSX band filter fluxes in Wm- can be calculated by multiplying the SED ΜΟΝ flux in Janskys by the bandwidth of the filter (4.041x1072000)., The MSX band filter fluxes in $^{-2}$ can be calculated by multiplying the SED MSX flux in Janskys by the bandwidth of the filter \citep[4.041~$\times$~10$^{-14}$.909 The weighted mea ratio of the total bolometric flux to the MSX band filter flux can therefore be obtained for each source. in à similar manner to that used for the bolometric flux (see refS:tests). with the uncertainty given by the weighted standard deviation.," The weighted mean ratio of the total bolometric flux to the MSX band filter flux can therefore be obtained for each source, in a similar manner to that used for the bolometric flux (see \\ref{S:tests}) ), with the uncertainty given by the weighted standard deviation."910 The results of these calculations do not appear to be dependent on source flux or type (see left-hand plot of Figure 7)). though the scatter in sources is relatively laree and so may mask such relationships.," The results of these calculations do not appear to be dependent on source flux or type (see left-hand plot of Figure \ref{F:results_nofir_ftotf21}) ), though the scatter in sources is relatively large and so may mask such relationships."911 The ratio itself has a roughly log-normal distribution. as shown in the right-hand plot of Figure 7..," The ratio itself has a roughly log-normal distribution, as shown in the right-hand plot of Figure \ref{F:results_nofir_ftotf21}."912 Sources with MIPSGAL data generally have a slightly smaller spread than the general distribution. so the properties of the Gaussian fits to the mean ratio for these 613 sources are used.," Sources with MIPSGAL data generally have a slightly smaller spread than the general distribution, so the properties of the Gaussian fits to the mean ratio for these 613 sources are used."913 This mean ratio was therefore used to obtain estimates of the bolometric flux of RMS sources which do not have far-IR data. since all have measured MSX fluxes.," This mean ratio was therefore used to obtain estimates of the bolometric flux of RMS sources which do not have far-IR data, since all have measured MSX fluxes."914" The bolometric fluxes obtained for sources with TIMMI2 (69) or GLIMPSE PSC (20) fluxes. as well as split sources (19). were apportioned using the method discussedin refS:data,imuni2."," The bolometric fluxes obtained for sources with TIMMI2 (69) or GLIMPSE PSC (20) fluxes, as well as split sources (19), were apportioned using the method discussedin \\ref{S:data_timmi2}."915".Anexampleo ftheseresultsispresentedinTable 2.. whereer doBAT) liver:= Autof mPa thistable.ana τν = τον|Veg savailableontineattheC""D DS ""n viaanonymo strasbg.fr(130.79.]25.5)0rvialittp/ledsweb.ustrasbg.frícgi—binlqeatJ/A+Af."," An example of these results is presented in Table \ref{T:nofir_results_full}, while the full version of this table is available online at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.125.5) or via http://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/."916 Estimate total fluxes were obtained using this method for an additional 280 young RMS sources., Estimate total fluxes were obtained using this method for an additional 280 young RMS sources.917 The uncertainty in the total flux for each source is a combination in quadrature of the uncertainty in Εμ / Faysyοι and the uncertainty in the MSX flux., The uncertainty in the total flux for each source is a combination in quadrature of the uncertainty in $F_{\rm{Bol}}$ $/$ $F_{MSX~21}$ and the uncertainty in the MSX flux.918 The fact that the Gaussian fit to Fao / Farsyoi is performed in log space results in asymmetric uncertainties. which are propagated through to the total flux.," The fact that the Gaussian fit to $F_{\rm{Bol}}$ $/$ $F_{MSX~21}$ is performed in log space results in asymmetric uncertainties, which are propagated through to the total flux."919 The distributions of luminosities for sources without far-IR data identified as YSO or region with uniquely assigned distances is shown in Figure 8.. with the distributions of SED derived luminosities from Figure 5 also shown for comparison.," The distributions of luminosities for sources without far-IR data identified as YSO or region with uniquely assigned distances is shown in Figure \ref{F:results_nofir_nplots}, with the distributions of SED derived luminosities from Figure \ref{F:discussion_nplots} also shown for comparison."920 Due to the lower number of sources (110 YSOs and 103 regions). the bin size for the distributions for luminosities derived using Fpe / Fagsxσι 18 twice that used for the SED lummosities.," Due to the lower number of sources (110 YSOs and 103 regions), the bin size for the distributions for luminosities derived using $F_{\rm{Bol}}$ $/$ $F_{MSX~21}$ is twice that used for the SED luminosities."921 The spread of sources is larger than for the SED lummosities. probably caused by the larger uncertainties 1n the bolometric fluxes derived using the MSX flux and the fact that these sources are in regions which are confused at far- wavelengths.," The spread of sources is larger than for the SED luminosities, probably caused by the larger uncertainties in the bolometric fluxes derived using the MSX flux and the fact that these sources are in regions which are confused at far-IR wavelengths."922 However the general shape of the distributions is similar. and the peaks in the distributions are near those for the SED luminosities.," However the general shape of the distributions is similar, and the peaks in the distributions are near those for the SED luminosities."923 This method of estimating the far-IR flux therefore seems to provide a reasonable estimate when no IR information ts available., This method of estimating the far-IR flux therefore seems to provide a reasonable estimate when no far-IR information is available.924" As with the ratio ofΕμ / Faysy2, discussed above. various other ratios both of filter fluxes to the bolometric flux and filter fluxes to other filter fluxes can be calculated from the model SED fits."," As with the ratio of$F_{\rm{Bol}}$ $/$ $F_{MSX~21}$ discussed above, various other ratios both of filter fluxes to the bolometric flux and filter fluxes to other filter fluxes can be calculated from the model SED fits."925 However the SED models of Robitailleetal.(2006) do not include PAH emission features. so the flux ratios for bands where these are important may be different. e.g. the PAH band may affect ratios involving the MSX band.," However the SED models of \citet[][]{Robitaille2006} do not include PAH emission features, so the flux ratios for bands where these are important may be different, e.g. the PAH band may affect ratios involving the MSX band."926 The results of these fits are presented in Table 3.., The results of these fits are presented in Table \ref{T:results_filters}. .927 In order to explore the relative accuracy of other methods commonly used to obtain bolometric fluxes compared to the SED fitter results. we performed calculations using both simple trapezium rule integration and a combination of two greybody fits (e.g.Minieretal...2005:Hillal..2009) with model rather than observed fluxes.," In order to explore the relative accuracy of other methods commonly used to obtain bolometric fluxes compared to the SED fitter results, we performed calculations using both simple trapezium rule integration and a combination of two greybody fits \citep[e.g.][]{Minier2005,Hill2009} with model rather than observed fluxes."928 The SED model ‘observed’ flux for each fit was first corrected for the extinetion derived by that fit., The SED model `observed' flux for each fit was first corrected for the extinction derived by that fit.929 Next the weighted mean and weighted standard deviation were calculated in order to obtain a corrected model flux with error for each filter where data were input to the fitter for that source., Next the weighted mean and weighted standard deviation were calculated in order to obtain a corrected model flux with error for each filter where data were input to the fitter for that source.930" The greybody fit was obtained using two temperature components given by: where 8& 1s the source size. T is the dust temperature. το 1s the optical depth at frequency v and f is the dust emissivity index such that 7 can be evaluated at any frequency relative to a reference frequency v,,;."," The greybody fit was obtained using two temperature components given by: where $\theta$ is the source size, T is the dust temperature, $\tau_{\nu}$ is the optical depth at frequency $\nu$ and $\beta$ is the dust emissivity index such that $\tau$ can be evaluated at any frequency relative to a reference frequency $\nu_{ref}$ ."931" A total of five free parameters were used for the fits: the temperatures corresponding to the peaks of the two greybody components at lower and higher temperature (T, and T>). the referenceoptical depth ἔτι) at um.. the source size in areseconds ofthe high-temperature component (#2) and Bj."," A total of five free parameters were used for the fits: the temperatures corresponding to the peaks of the two greybody components at lower and higher temperature $_{1}$ and $_{2}$ ), the referenceoptical depth $\tau_{ref}$ ) at , the source size in arcseconds ofthe high-temperature component $\theta_{2}$ ) and $\beta_{1}$ ."932 B» was kept constant at 1. while 8. the," $\beta_{2}$ was kept constant at 1, while $\theta_{1}$ , the"933The discovery of an accelerated: cosmic expansion has become one of the biggest. puzzles in modern cosmology over the last. 10 vears.,The discovery of an accelerated cosmic expansion has become one of the biggest puzzles in modern cosmology over the last 10 years.934 Several scientific probes have been proposed to understand the nature of this acceleration., Several scientific probes have been proposed to understand the nature of this acceleration.935" From. ""eeomietrical"" tests based on measurements of the distance-redshift relation such as barvon acoustic oscillations (D.XO) or ‘Type la supernovae. to “growth” tests sensitive to the erowth rate of perturbations such as redshift space distortions. (RSD). weak lensing or cluster abundance."," From “geometrical” tests based on measurements of the distance-redshift relation such as baryon acoustic oscillations (BAO) or Type Ia supernovae, to “growth” tests sensitive to the growth rate of perturbations such as redshift space distortions (RSD), weak lensing or cluster abundance."936 The success of these probes relies in the implementation of massive. and many times cdecicated. observational campaigns that will scan a good fraction of the observable Universe.," The success of these probes relies in the implementation of massive, and many times dedicated, observational campaigns that will scan a good fraction of the observable Universe."937 Some such surveys will base their science in galaxy recshifts derived: spectroscopically. what) provides accurate radial positions.," Some such surveys will base their science in galaxy redshifts derived spectroscopically, what provides accurate radial positions."938 Others will instead measure redshift’ photometrically., Others will instead measure redshift photometrically.939 “Phis vields poorer determination of racial positions but allows to go deeper in redshift and have higher sampling rate., This yields poorer determination of radial positions but allows to go deeper in redshift and have higher sampling rate.940" The later group involves the Dark Energy (DES). the Physics of the Accclerating Universe (PAU) and the Panoramic Survey ""Telescope and Hapid Response (PanStarrs) as well as proposals such as the Large Svnoptic Survey (LSS'T) and the imaging component of ESA/Euclid survey."," The later group involves the Dark Energy (DES), the Physics of the Accelerating Universe (PAU) and the Panoramic Survey Telescope and Rapid Response (PanStarrs) as well as proposals such as the Large Synoptic Survey (LSST) and the imaging component of ESA/Euclid survey."941 Perhaps the most exciting results related to the large scale structure of the Universe to date have been obtained using spectroscopic data from surveys such as the two degree, Perhaps the most exciting results related to the large scale structure of the Universe to date have been obtained using spectroscopic data from surveys such as the two degree942field. for which a finer resolution is needed.,"field, for which a finer resolution is needed."943 A galaxy with a distance uncertainty of αἱ 10 Alpe gives a peculiar radial velocity uncertainty of something over 60 km |. which is about the largest which could be tolerated here.," A galaxy with a distance uncertainty of at 10 Mpc gives a peculiar radial velocity uncertainty of something over 60 km $^{-1}$, which is about the largest which could be tolerated here."944 This means (hat some popular methods. such as the Tully-Fisher relation. are not useful. and in fact almost. all distances were obtained using Cepheid variables (Ceph in the table) or the brightness of the tip of the Red Giant Branch as found in a 7. V—£ color-amnagnitude diagram (TRGB).," This means that some popular methods, such as the Tully-Fisher relation, are not useful, and in fact almost all distances were obtained using Cepheid variables (Ceph in the table) or the brightness of the tip of the Red Giant Branch as found in a $I$, $V-I$ color-magnitude diagram (TRGB)."945 Some variations on the latter use different fillers (A. the SDSS svstem). or a sliehtlv dillerent calibration for the absolute magnitude of the tip (M;=—4.05 instead ol -4.00): thev are marked. where they occur. and any. resulting differences in distance lie within the «quoted errors.," Some variations on the latter use different filters $K$, the SDSS system), or a slightly different calibration for the absolute magnitude of the tip $M_I = -4.05$ instead of -4.00); they are marked where they occur, and any resulting differences in distance lie within the quoted errors."946 The technique of surface brightness [uctuation (SBF) seems «quite promising and should have added: a few more objects., The technique of surface brightness fluctuation (SBF) seems quite promising and should have added a few more objects.947 However. the SBF distances for DDO 181 (Hidalgoοἱal. and ESO 540-32 (Jerjenοἱal.1993) disagree rather strongly with the TRGB distances (Ixarachentsevοἱal.(20020). and Jerejen&Rejkuba (2001).. respectively). and NGC 4736 is not much better (Tonryetal.(2001) ancl Ixaraclientsevetal. (2003a))).," However, the SBF distances for DDO 181 \citep{HMA03} and ESO 540-32 \citep{JFB98} disagree rather strongly with the TRGB distances \citet{KSM02} and \citet{JR01}, , respectively), and NGC 4736 is not much better \citet{T01} and \citet{KSD03}) )."948 Since the number of galaxies with only SBF distances is small. it was felt better to leave them out rather than add a source of unknown error.," Since the number of galaxies with only SBF distances is small, it was felt better to leave them out rather than add a source of unknown error."949 Where more than one distance estimate was available. and all were consistent. (μον were combined.," Where more than one distance estimate was available, and all were consistent, they were combined."950 If they were within their stated relative errors. (μον were averaged ancl the uncertainty reduced by a factor of WN: if they were between one and (wo uncertainties distant. they were averaged and the larger uncertainty taken.," If they were within their stated relative errors, they were averaged and the uncertainty reduced by a factor of $\sqrt{N}$; if they were between one and two uncertainties distant, they were averaged and the larger uncertainty taken."951 If they were not reconcilable (for instance. with IC 10) a judgement as to the more reliable distance(s) was mace.," If they were not reconcilable (for instance, with IC 10) a judgement as to the more reliable distance(s) was made."952 Even though 149 is a gratilving number of galaxies to work with (being more than four limes (he number of good data-points used in a previous calculation. Whiting. (2003))). it still does not include half of the total number present: nor. more importantly. are all the brightest galaxies (here.," Even though 149 is a gratifying number of galaxies to work with (being more than four times the number of good data-points used in a previous calculation, \citet{WH03}) ), it still does not include half of the total number present; nor, more importantly, are all the brightest galaxies there."953 To include all galaxies which have been known or suspected to be brighter than Mj~18.5 and in the Local Volume. 21 additional objects are presented in Table (2)).," To include all galaxies which have been known or suspected to be brighter than $M_B \sim 18.5$ and in the Local Volume, 21 additional objects are presented in Table \ref{data2}) )."954 The methods used in several of these cases might in fact eive high-quality. data. since such (things as the brightest-star method and SBF seem (o work satisfactorily with larger galaxies.," The methods used in several of these cases might in fact give high-quality data, since such things as the brightest-star method and SBF seem to work satisfactorily with larger galaxies."955 This needs to be checked. though. and Table (2)) could usefully be taken as a target list lor a deep TRGB observing program.," This needs to be checked, though, and Table \ref{data2}) ) could usefully be taken as a target list for a deep TRGB observing program."956 Brightest-star distances are assigned an uncertainty of 0.4 magnitude., Brightest-star distances are assigned an uncertainty of 0.4 magnitude.957 Distances to the major Sculptor Group galaxies were determined by Puche&Carignan(1988). [roni a variety ol methods: as (hose determined since lor NGC 300. 253 and 7793 are svstematically larger. those lor NGC 55 and 247 have been adjusted accordingly (note that allof these are within," Distances to the major Sculptor Group galaxies were determined by \citet{PC88} from a variety of methods; as those determined since for NGC 300, 253 and 7793 are systematically larger, those for NGC 55 and 247 have been adjusted accordingly (note that allof these are within"958projected onto S and thence mapped to C.* and the resulting distributions were smoothed with an iterated boxcar.,"projected onto ${\bf S}$ and thence mapped to ${\bf C}$, and the resulting distributions were smoothed with an iterated boxcar."959 In these images. color is used to indicate radii. with erev for the nuclei. blue for small racii. green and vellow for intermediate racii. and red for large radii.," In these images, color is used to indicate radii, with grey for the nuclei, blue for small radii, green and yellow for intermediate radii, and red for large radii."960 As Figure 7 shows. the remnants have cliverse kinematics.," As Figure \ref{fig07} shows, the remnants have diverse kinematics."961 Mlany of the structures seen here can be identified with components celineatecl in Figure 6.., Many of the structures seen here can be identified with components delineated in Figure \ref{fig06}.962. Gas in the nuclei. shown in grew. produces the broadest distributions. but each nucleus has a well-defined net. spin direction which is marked by α 7N7 in these plots.," Gas in the nuclei, shown in grey, produces the broadest distributions, but each nucleus has a well-defined net spin direction which is marked by a “N” in these plots."963" Likewise. the “BvD? and ""L7 symbols mark the net spin directions of the bars. disks. and loops: these coincide with the angular momenta of gas particles. shown in blue and green."," Likewise, the “B”, “D”, and “L” symbols mark the net spin directions of the bars, disks, and loops; these coincide with the angular momenta of gas particles, shown in blue and green."964 On the other hand. the spin vectors of the disks. labeled 7] and 727 generally don't. correspond. to. favored. spin directions in the remnants. but often appear associated with material in tidal tails. here shown in vellow and red.," On the other hand, the spin vectors of the disks, labeled “1” and “2”, generally don't correspond to favored spin directions in the remnants, but often appear associated with material in tidal tails, here shown in yellow and red."965 A few remnants displav fairly. simple structures., A few remnants display fairly simple structures.966 For example. remnants of Dlltect. encounters tend. to. have nuclei and disks with reasonably well-aligned: rotation. basically because the spins and orbital motion of the initial encounter reinforce. cach another.," For example, remnants of DIRect encounters tend to have nuclei and disks with reasonably well-aligned rotation, basically because the spins and orbital motion of the initial encounter reinforce each another."967 But in many cases the eas has some kind of kincmatic misalignment., But in many cases the gas has some kind of kinematic misalignment.968 All remnants of RETroerade passages have disjoint angular momenta clistributions. with counter-rotating nucle: and disks.," All remnants of RETrograde passages have disjoint angular momenta distributions, with counter-rotating nuclei and disks."969 Vhis is not surprising given the rather contrived. initial conditions of these mergers: the distribution of gas between direct and retrograde: rotation rellects the competition of spin ancl orbital angular momenta in these encounters., This is not surprising given the rather contrived initial conditions of these mergers; the distribution of gas between direct and retrograde rotation reflects the competition of spin and orbital angular momenta in these encounters.970 Several other remnants have gas nuclei with dramatic kinematic misalignments: remnant POL 1:1 € has a counter-rotating nucleus. while remnants INC 1:1 € and D have nuclei which rotate about axes roughly. perpendicular to the axes of their clisks.," Several other remnants have gas nuclei with dramatic kinematic misalignments: remnant POL 1:1 C has a counter-rotating nucleus, while remnants INC 1:1 C and D have nuclei which rotate about axes roughly perpendicular to the axes of their disks."971 As noted above. many remnant disks have warps.," As noted above, many remnant disks have warps."972 A disk which warps progressively along a single direction appears as a roughly linear feature with a smooth color gradient in ligure 7.. as in the plot for remnant INC 1:1: €. In other disks the warp's direction changes with radius. perhaps as à result of differential precession: such warps produce curved forms as in the plot for remnant POL 1:1 C. or even circular features like the one in RET 1:1. D. Most. of these. disks are fairly [at at small μασ. as if the inner cisks are locked into a single plane bv self-gravitv.," A disk which warps progressively along a single direction appears as a roughly linear feature with a smooth color gradient in Figure \ref{fig07}, as in the plot for remnant INC 1:1 C. In other disks the warp's direction changes with radius, perhaps as a result of differential precession; such warps produce curved forms as in the plot for remnant POL 1:1 C, or even circular features like the one in RET 1:1 D. Most of these disks are fairly flat at small radii, as if the inner disks are locked into a single plane by self-gravity."973 Warps in the outer clisks of these remnants may be excited. by the return of tail material with misaligned. angular momentum., Warps in the outer disks of these remnants may be excited by the return of tail material with misaligned angular momentum.974 In Figure 3 these remnants exhibit fairly smooth transitions from the disks (blue or green) to the tails (vellow or red): thus the angular momentum of the gas changes in a continuous manner with radius., In Figure \ref{fig07} these remnants exhibit fairly smooth transitions from the disks (blue or green) to the tails (yellow or red); thus the angular momentum of the gas changes in a continuous manner with radius.975 The tails contain large amounts of angular momentum: as this material falls back it could exert strong torques. creating long-lived warps in remnant disks.," The tails contain large amounts of angular momentum; as this material falls back it could exert strong torques, creating long-lived warps in remnant disks."976 Remnant Dlh 1:1 € provides the example of a precessing inner disk shown in Figure S.., Remnant DIR 1:1 C provides the example of a precessing inner disk shown in Figure \ref{fig08}.977 This disk. bounded by the dottedline at logr21 in the upper-Ieft. panel of Figure 6.. begins forming as soon as the nuclei of its »ogenitors have merged and the gravitational potential has settled down.," This disk, bounded by the dottedline at $\log r \simeq -1.2$ in the upper-left panel of Figure \ref{fig06}, begins forming as soon as the nuclei of its progenitors have merged and the gravitational potential has settled down."978 By /=4.5 its seen roughly edge-on extending roni eight to two oclock: it has precessed counter-clockwise w 9907 by f=6., By $t = 4.5$ it's seen roughly edge-on extending from eight to two o'clock; it has precessed counter-clockwise by $\sim 90^\circ$ by $t = 6$.979 During this time the inner clisk remains relatively well-aligned. with the nucleus., During this time the inner disk remains relatively well-aligned with the nucleus.980 But between the nucleus and the innerdisk lies a ring of material. inclined wo 1307.2 which precesses by ~1207 between f=4.5 and 6.," But between the nucleus and the innerdisk lies a ring of material, inclined by $\sim 130^\circ$, which precesses by $\sim 120^\circ$ between $t = 4.5$ and $6$."981 UW the nucleus ancl inner disk are actually ocked in alignment. they must. be coupled. &ravitationallv. since the ring lies between them.," If the nucleus and inner disk are actually locked in alignment, they must be coupled gravitationally, since the ring lies between them."982 ‘To rough approximation. both the inner disk and extra-nuclear ring in this remnant behave like inclined. rotators oecessing in an oblate potential.," To rough approximation, both the inner disk and extra-nuclear ring in this remnant behave like inclined rotators precessing in an oblate potential."983 But the middle row in Figure S. provides evidence that viscous forces also influence he dynamics of the inner disk., But the middle row in Figure \ref{fig08} provides evidence that viscous forces also influence the dynamics of the inner disk.984 Phese images show the gas xwticles weighted by their dissipation rate fe: the inner disk is quite prominent. as is the gas just bevond the inner disk.," These images show the gas particles weighted by their dissipation rate $\dot{u}$; the inner disk is quite prominent, as is the gas just beyond the inner disk."985 The latter exhibits a spiral dissipation pattern. implying hat angular momentum is being transported racially.," The latter exhibits a spiral dissipation pattern, implying that angular momentum is being transported radially."986 One consequence of this viscous coupling may. be the gradual decrease in the inclination of the inner disk., One consequence of this viscous coupling may be the gradual decrease in the inclination of the inner disk.987 Phe inner disk of DIR OC acquired. its tilt in a fairly straightforward manner., The inner disk of DIR 1:1 C acquired its tilt in a fairly straightforward manner.988 Some 74 percent. of the gas particles in this component came from the ;/=71 galaxy. and the angular momentum of these particles determines the initial orientation of the disk.," Some $74$ percent of the gas particles in this component came from the $i = 71^\circ$ galaxy, and the angular momentum of these particles determines the initial orientation of the disk."989 Some misaligned components in other remnants may be explained in a similar way: in these cases the misaligned material comes preferentially from one progenitor or the other., Some misaligned components in other remnants may be explained in a similar way; in these cases the misaligned material comes preferentially from one progenitor or the other.990 But the relationship between the originating disk and the final spin direction is not always simple — gravitational ancl hvedrodsnamical torques play a large role in determining the final rotation direction of the gas., But the relationship between the originating disk and the final spin direction is not always simple – gravitational and hydrodynamical torques play a large role in determining the final rotation direction of the gas.991 The simulations cover only the carly evolution. of. these merger remnants: they include the violent dissipative events which form the inner parts of gas disks but end. shortly thereafter., The simulations cover only the early evolution of these merger remnants; they include the violent dissipative events which form the inner parts of gas disks but end shortly thereafter.992 On longer time-scales. much of the gas lingcring in tidal tails will fall back (Ilibbard Alihos 1995). evolve through loop-like structures. ancl finally seek. out. closed orbits.," On longer time-scales, much of the gas lingering in tidal tails will fall back (Hibbard Mihos 1995), evolve through loop-like structures, and finally seek out closed orbits."993 Hs computationally expensive to follow this process much past the stage shown here. but counting the bound tail gas. it appears that the DIItect and RieErogracde remnants have enough material to form disks containing up to 25 percent of the total gas. while the POLar and. INClined mergers can form disks containing 25 to nearly 60 percent.," It's computationally expensive to follow this process much past the stage shown here, but counting the bound tail gas, it appears that the DIRect and RETrograde remnants have enough material to form disks containing up to $25$ percent of the total gas, while the POLar and INClined mergers can form disks containing $25$ to nearly $60$ percent."994 Phe time-scale for this infall is casily estimated in the limit where the tail gas is just barely bound to the remnant., The time-scale for this infall is easily estimated in the limit where the tail gas is just barely bound to the remnant.995 1n this limit the tail material moves on nearly radial orbits in an approximately Ixeplerian potential., In this limit the tail material moves on nearly radial orbits in an approximately Keplerian potential.996 Suppose that a tail is formed during a pericentric passage at time frei., Suppose that a tail is formed during a pericentric passage at time $t_{\rm peri}$ .997" A parcel of tail material with specific binding energy. Lo<0 reaches apogalacticon at radius raj;&CAL,|MS E]. and. falls back at time"," A parcel of tail material with specific binding energy $E < 0$ reaches apogalacticon at radius $r_{\rm apo} \simeq G (M_1 \! + \! M_2) / |E|$ , and falls back at time"998"from the bright galaxies). but instead trace the gas we would expect <$,,> in the Fourier plane to have the same shape as T. rather than being constant.","from the bright galaxies), but instead trace the gas we would expect $<S_{pt}>$ in the Fourier plane to have the same shape as $\tilde{T}$, rather than being constant."999 For simplicity. we assume point sources are strongly centrally concentrated and place them at the cluster centers.," For simplicity, we assume point sources are strongly centrally concentrated and place them at the cluster centers."1000 In figure 3. we show the SZ power spectrum that would be inferred., In figure \ref{fig:cl_pts_2} we show the SZ power spectrum that would be inferred.1001 The power spectrum due to the point sources alone has been subtracted., The power spectrum due to the point sources alone has been subtracted.1002 Unsubtracted point sources cause a significant underestimate of the SZ angular power spectrum., Unsubtracted point sources cause a significant underestimate of the SZ angular power spectrum.1003 Assuming a larger contamination of of the SZ flux for a z20.5 I0?47!M eliminates nearly all of the SZ power. while dropping the contamination.. to only reduces the missing power to only about20%.," Assuming a larger contamination of of the SZ flux for a $z=0.5$ $10^{15}h^{-1}M_\odot$ eliminates nearly all of the SZ power, while dropping the contamination to only reduces the missing power to only about."1004. Point source contamination makes interpretation of detection of fluctuations in the CMB from the thermal SZ effect very difficult., Point source contamination makes interpretation of detection of fluctuations in the CMB from the thermal SZ effect very difficult.1005 To compare these fluctuations to predictions from either semi-analytic modeling or numerical simulations. some treatment of the effects of point source subtraction (or non-subtraction) is required.," To compare these fluctuations to predictions from either semi-analytic modeling or numerical simulations, some treatment of the effects of point source subtraction (or non-subtraction) is required."1006 This would require a recipe for the cluster radio galaxy populations., This would require a recipe for the cluster radio galaxy populations.1007 Current predictions of the thermal SZ power spectrum are almost certainly overestimating the thermal SZ power that CBI could observe., Current predictions of the thermal SZ power spectrum are almost certainly overestimating the thermal SZ power that CBI could observe.1008 The measured power at 30 GHz at high multipoles could be less than of the true SZ power. but more likely is measuring roughly50-75%.," The measured power at 30 GHz at high multipoles could be less than of the true SZ power, but more likely is measuring roughly."1009".. Specifically. the recent tentative report of temperature fluctuations at high multipoles of close to 500pK (Masoneraf,2002).. if due to the thermal SZ effect would indicate a true signal on the sky of roughly [0ὔθμκ”."," Specifically, the recent tentative report of temperature fluctuations at high multipoles of close to $500 \mu K^2$ \citep{mason02}, if due to the thermal SZ effect would indicate a true signal on the sky of roughly $1000 \mu K^2$."1010 A temperature of 25j/K represents a bit of a challenge for theoretical models (Bonderαἱ.2002:KomatsuandSel-jak 2002).. so a true signal greater than 30j/K would require a major rethinking of the physics of galaxy clusters and/or some fine tuning of cosmological parameters.," A temperature of $25 \mu K$ represents a bit of a challenge for theoretical models \citep{bond02,komatsu02}, so a true signal greater than $30 \mu K$ would require a major rethinking of the physics of galaxy clusters and/or some fine tuning of cosmological parameters."1011 The simplest way to increase the expected SZ power is to increase σε., The simplest way to increase the expected SZ power is to increase $\sigma_8$.1012 A true SZ signal of more than 30//K would suggest os—1.1 (KomatsuandSeljak 2002).. a value that is not preferred by current CMB data (Bondetαἱ.2002) but is not ruled out.," A true SZ signal of more than $30 \mu$ K would suggest $\sigma_8 \ga 1.1$ \citep{komatsu02}, a value that is not preferred by current CMB data \citep{bond02} but is not ruled out."1013 From figure 3 of Bonderal...(2002) such a high value could most easily be accommodated 1f the Hubble constant were significantly lower than the value suggested by HST measurements (Freedmanetal. 2001)., From figure 3 of \citet{bond02} such a high value could most easily be accommodated if the Hubble constant were significantly lower than the value suggested by HST measurements \citep{freedman01}.1014. Alternatively. rf the measured power at high { really is a measurement of the thermal SZ effect. this would be evidence that low-mass clusters at z0.5 are remarkably devoid of bright radio sources.," Alternatively, if the measured power at high $\ell$ really is a measurement of the thermal SZ effect, this would be evidence that low-mass clusters at $z\sim0.5$ are remarkably devoid of bright radio sources."1015 This suggests that upcoming SZ surveys at 30 GHz (SZA) or 15 GHz (AMI) should have surprisingly clear extragalactic skies., This suggests that upcoming SZ surveys at 30 GHz (SZA) or 15 GHz (AMI) should have surprisingly clear extragalactic skies.1016 The statistics of point sources at high radio frequencies are very poorly constrained. making detailed predictions of point source contamination difficult.," The statistics of point sources at high radio frequencies are very poorly constrained, making detailed predictions of point source contamination difficult."1017 We have adopted an approach to modeling point source contamination of SZ signal that is empirically motivated. with only the relative importance of point sources to SZ signal at a single mass scale as a free parameter.," We have adopted an approach to modeling point source contamination of SZ signal that is empirically motivated, with only the relative importance of point sources to SZ signal at a single mass scale as a free parameter."1018 Upcoming SZ/CMB experiments with high angular resolution. specifically for the purpose of point source detection. such as AMI and SZA will provide a wealth of information on radio point sources at these frequencies. while at the same time providing valuable information on the SZ effect from galaxy clusters.," Upcoming SZ/CMB experiments with high angular resolution, specifically for the purpose of point source detection, such as AMI and SZA will provide a wealth of information on radio point sources at these frequencies, while at the same time providing valuable information on the SZ effect from galaxy clusters."1019 Single frequency measurements of the fluctuations due to the thermal SZ effect at frequencies below ~90 GHz that do not subtract point sources with a small beam to a fairly low flux level will be contaminated at a largely unknown but almost certainly significant level., Single frequency measurements of the fluctuations due to the thermal SZ effect at frequencies below $\sim$ 90 GHz that do not subtract point sources with a small beam to a fairly low flux level will be contaminated at a largely unknown but almost certainly significant level.1020 Higher frequency measurements could have similar problems from dusty starburst galaxies. but most of these sources are expected to not be associated with the galaxy cluster members.," Higher frequency measurements could have similar problems from dusty starburst galaxies, but most of these sources are expected to not be associated with the galaxy cluster members."1021 While lensing effects lead to an enhancement of the confusion noise (Blain1998).. it does not lead. on average to an increased average flux.," While lensing effects lead to an enhancement of the confusion noise \citep{blain98}, it does not lead, on average to an increased average flux."1022 Experiments with multiple frequencies will be required for a robust determination of the amplitude of the thermal SZ signal. and the strong correlations between radio point sources and galaxy clusters and submm point sources and galaxy clusters (primarily due to gravitational lensing) will require careful attention.," Experiments with multiple frequencies will be required for a robust determination of the amplitude of the thermal SZ signal, and the strong correlations between radio point sources and galaxy clusters and submm point sources and galaxy clusters (primarily due to gravitational lensing) will require careful attention."1023 The general problem of correlation between secondary anisotropies and foregrounds will be increasingly important., The general problem of correlation between secondary anisotropies and foregrounds will be increasingly important.1024 The thermal SZ effect may be the most significant example. but correlations between the lensing of the CMB. for example. and radio and submm point sources will reduce the expected signal and/or modify the noise properties of any attempted reconstructions.," The thermal SZ effect may be the most significant example, but correlations between the lensing of the CMB, for example, and radio and submm point sources will reduce the expected signal and/or modify the noise properties of any attempted reconstructions."1025" Clearly. a better understanding of the covariance between various foregrounds and backgrounds and secondary anisotropies of interest will be required if such signals are to be used as useful tests of our understanding of cosmology and structure formation,"," Clearly, a better understanding of the covariance between various foregrounds and backgrounds and secondary anisotropies of interest will be required if such signals are to be used as useful tests of our understanding of cosmology and structure formation."1026some of the correlations discussed in our study occur in other galaxies. such as the correlation between (he presence of an LAINB ancl the metallicity (color is the proxy [or metallidtv in most extragalactic studies. aside [rom Local Group galaxies: e.g.. &Zepf 20023).,"Some of the correlations discussed in our study occur in other galaxies, such as the correlation between the presence of an LMXB and the metallicity (color is the proxy for metallicity in most extragalactic studies, aside from Local Group galaxies; e.g., \citealt{kund02}) )."1027 Yet other galaxies will permit the studies of the LMXD rate with age. since some galaxies are vounger than others or have globular clusters known to be voung (e.g.. M31. LMC).," Yet other galaxies will permit the studies of the LMXB rate with age, since some galaxies are younger than others or have globular clusters known to be young (e.g., M31, LMC)."1028 With Galactic globular clusters. it might be possible to measure the binary [fraction rate for high and low metallicity svstems. testing our conclusion.," With Galactic globular clusters, it might be possible to measure the binary fraction rate for high and low metallicity systems, testing our conclusion."1029 Also. the number of globular clusters with LAINBs is a modest 12 svstems. and the study of certain elliptical galaxies offers (he opportunity of increasing (he statistics by an order of magnitude (e.g.. Angelini.Loewenstein.&Mushotzkv. 2001: Irwin.Athev.&Bregman 2003)). since rj can be measured in external galaxies al the distance of the Virgo Cluster 2005).," Also, the number of globular clusters with LMXBs is a modest 12 systems, and the study of certain elliptical galaxies offers the opportunity of increasing the statistics by an order of magnitude (e.g., \citealt{ang01}; ; \citealt{irwin03}) ), since $r_h$ can be measured in external galaxies at the distance of the Virgo Cluster \citep{jord05}."1030. Finally. we look forward to the improvement in models. which in principle should be able to predict the frequency of LMXDs for elobular clusters of various initial properties ancl evolutionary states.," Finally, we look forward to the improvement in models, which in principle should be able to predict the frequency of LMXBs for globular clusters of various initial properties and evolutionary states."1031 We would like to acknowledge support for this work from through grant., We would like to acknowledge support for this work from through grant NAG5-10765.1032 Also. we would like to thank Mario Mateo. Renato Dupke. Ecward Llove-Davies.aud an anonymous referee for their valuable aclvice.," Also, we would like to thank Mario Mateo, Renato Dupke, Edward Lloyd-Davies,and an anonymous referee for their valuable advice."1033Figure 6. shows the distribution in ία. e) space of ov experiments. in the same format as Figs.,"Figure \ref{fig:55a} shows the distribution in $a$, $e$ ) space of our experiments, in the same format as Figs."1034 3 and 5.. with Poisson errors of ~ per bin.," \ref{fig:37b} and \ref{fig:38a}, , with Poisson errors of $\sim$ per bin."1035 We see three local maxima: 1) a relatively narrow maximum at à  1.0 AU. e ~ 0.03. 2) à broad maximum centered roughly at ¢ ~ 2.0 AU. e ~ 0.08 but which extends to higher values of a. and 3) & ~ 3 AU. e ~ 0.17.," We see three local maxima: 1) a relatively narrow maximum at $a$ $\sim$ 1.0 AU, $e$ $\sim$ 0.03, 2) a broad maximum centered roughly at $a$ $\sim$ 2.0 AU, $e$ $\sim$ 0.08 but which extends to higher values of $a$, and 3) $a$ $\sim$ 3 AU, $e$ $\sim$ 0.17."1036 Region | is of great astrobiological interest. as it lies in the habitable zone of its parent star. which is bounded by the black dashed lines.," Region 1 is of great astrobiological interest, as it lies in the habitable zone of its parent star, which is bounded by the black dashed lines."1037 Region 3 1s bordered by the 3:1 (2.84 AU) and 5:2 (3.2 AU) mean motion resonances with the outer planet., Region 3 is bordered by the 3:1 (2.84 AU) and 5:2 (3.2 AU) mean motion resonances with the outer planet.1038 We see no clear trend of survival rate with mean anomaly near these resonances., We see no clear trend of survival rate with mean anomaly near these resonances.1039 Paper | found that no test particles survived in this system for longer than | Myr., Paper 1 found that no test particles survived in this system for longer than 1 Myr.1040 The region in which they survived the longest was for a between 0.5 and 1.5 AU at relatively low eccentricities., The region in which they survived the longest was for $a$ between 0.5 and 1.5 AU at relatively low eccentricities.1041 The 1:5 mean motion resonance (with the inner planet) is at 0.82 AU and the 5:1 resonance (with the outer planet) at 1.3 AU are located at the outskirts of the region we investigate., The 1:5 mean motion resonance (with the inner planet) is at 0.82 AU and the 5:1 resonance (with the outer planet) at 1.3 AU are located at the outskirts of the region we investigate.1042 Therefore. only very high order mean motion resonances are found in the center.," Therefore, only very high order mean motion resonances are found in the center."1043 We find no evidence of secular resonances in the region., We find no evidence of secular resonances in the region.1044 We performed 600 integrations of Saturn-mass planets in this system with c in the above mentioned region. and e between 0 and 0.2.," We performed 600 integrations of Saturn-mass planets in this system with $a$ in the above mentioned region, and $e$ between 0 and 0.2."1045 Of these 600 Saturns. 296 (49%)) survived for 100 Myr.," Of these 600 Saturns, 296 ) survived for 100 Myr."1046 Figure 7. shows the distribution of the surviving planets in these simulations., Figure \ref{fig:74a} shows the distribution of the surviving planets in these simulations.1047 We see three small islands of stability at (4.6) ~: 1) (1.0 AU. 0.02). 2) (1.0 AU. 0.1). and 3) (1.2 AU. 0.13).," We see three small islands of stability at $a,e$ ) $\simeq$: 1) (1.0 AU, 0.02), 2) (1.0 AU, 0.1), and 3) (1.2 AU, 0.13)."1048 These three islands lie at a slightly higher survival rate than the surrounding. larger region of stability between 0.9 - 1.2 AU with e< 0.15. in which the survival rate is75%.," These three islands lie at a slightly higher survival rate than the surrounding, larger region of stability between 0.9 - 1.2 AU with $e1049\leq$ 0.15, in which the survival rate is."1050. We see a strong trend in the survival rate of planets as a function of semimajor axis. as shown in Fig. 8..," We see a strong trend in the survival rate of planets as a function of semimajor axis, as shown in Fig. \ref{fig:74b}."1051 The fraction of systems which are stable for 100 Myr increases sharply between 0.8 and 1.0 AU. then flattens off and decreases slightly past 1.2 AU.," The fraction of systems which are stable for 100 Myr increases sharply between 0.8 and 1.0 AU, then flattens off and decreases slightly past 1.2 AU."1052 The stable zones found in Fig., The stable zones found in Fig.1053 7. lie at the peak of the curve., \ref{fig:74a} lie at the peak of the curve.1054 Menou Tabachnik (2003: hereafter MT) investigated the possibility of Earth-sized planets residing in the habitable zones (HZs) of known extrasolar planetary systems., Menou Tabachnik (2003; hereafter MT) investigated the possibility of Earth-sized planets residing in the habitable zones (HZs) of known extrasolar planetary systems.1055 The location of the HZ is a function of the luminosity (and therefore mass) of the host star. as well as the atmospheric composition of the planet (Kasting 1993).," The location of the HZ is a function of the luminosity (and therefore mass) of the host star, as well as the atmospheric composition of the planet (Kasting 1993)."1056 For each system MT integrated the orbits of 100 massless test particles in the HZ for 10° years., For each system MT integrated the orbits of 100 massless test particles in the HZ for $^6$ years.1057 They considered all four of our systems., They considered all four of our systems.1058 The HZs for each system are as follows - HD37124: 0.6-1.2 AU. HD38329: 1.4-3 AU. HD74156: 0.6-1.2 AU. and 55Cne: 0.7-1.3 AU.," The HZs for each system are as follows – HD37124: 0.6-1.2 AU, HD38529: 1.4-3 AU, HD74156: 0.6-1.2 AU, and 55Cnc: 0.7-1.3 AU."1059 MT found no surviving planets in the HZ of HD37124., MT found no surviving planets in the HZ of HD37124.1060 Their stability criterion requires a particle to remain in the HZ at all times. limiting its eccentricity such that the particle's aphelion and perihelion remain in the HZ.," Their stability criterion requires a particle to remain in the HZ at all times, limiting its eccentricity such that the particle's aphelion and perihelion remain in the HZ."1061 Paper | used over 500 test particles to systematically map out the region in HD37124 which is stable for test particles. finding it to be centered at 1 AU.," Paper 1 used over 500 test particles to systematically map out the region in HD37124 which is stable for test particles, finding it to be centered at 1 AU."1062 The eccentricities in this stable region are small enough to keep test particles in the HZ of the system throughout their orbits., The eccentricities in this stable region are small enough to keep test particles in the HZ of the system throughout their orbits.1063 In addition. we find three local maxima of the survival rate Saturn-mass planets in this system. all of whose orbits remain in the HZ.," In addition, we find three local maxima of the survival rate Saturn-mass planets in this system, all of whose orbits remain in the HZ."1064 For HD38529 our results are consistent with MT. as the stable region from Paper | lies well outside the HZ. and the region we investigated with Saturns does not overlap with the HZ.," For HD38529 our results are consistent with MT, as the stable region from Paper 1 lies well outside the HZ, and the region we investigated with Saturns does not overlap with the HZ."1065 In the case of 55Cne our results are again consistent with MT. who find that a significant fraction of low-inclination test particles survive at 1.0 AU. with eccentricities centered on 0.09.," In the case of 55Cnc our results are again consistent with MT, who find that a significant fraction of low-inclination test particles survive at 1.0 AU, with eccentricities centered on 0.09."1066 The stable region for 55Cne from Paper | encompasses the HZ entirely for eccentricities below 0.25., The stable region for 55Cnc from Paper 1 encompasses the HZ entirely for eccentricities below 0.25.1067 In addition. Table 3 shows a maximum in the survival rate of Saturns at (a. e) = (1.0 AU. 0.03). very close to the value from MT.," In addition, Table 3 shows a maximum in the survival rate of Saturns at $a$, $e$ ) = (1.0 AU, 0.03), very close to the value from MT."1068 MT's results for HD74156 are consistent with Paper |. but we have found two regions in the HZ which are stable for Saturn-mass planets in of cases.," MT's results for HD74156 are consistent with Paper 1, but we have found two regions in the HZ which are stable for Saturn-mass planets in of cases."1069 However. this may be due to the fact that the orbital elements used by MT are different than those we have used here.," However, this may be due to the fact that the orbital elements used by MT are different than those we have used here."1070 In particular. the semimajor axis of the outer planet used here is 0.35 AU larger (3.82 AU vs 3.47 AU). increasing the separation of the two giant planets and therefore possibly causing the region in between to become more stable for an additional companion.," In particular, the semimajor axis of the outer planet used here is 0.35 AU larger (3.82 AU vs 3.47 AU), increasing the separation of the two giant planets and therefore possibly causing the region in between to become more stable for an additional companion."1071 Note that the current value for HD74156ec is 3.40 AU (Naef 2004)., Note that the current value for HD74156c is 3.40 AU (Naef 2004).1072 Dvorak (2003) investigated the possibility of an unseen planet in HD74156. using both test particles and massive ones.," Dvorak (2003) investigated the possibility of an unseen planet in HD74156, using both test particles and massive ones."1073 They find a broad. relatively stable region for test particles between 0.9 and 1.4 AU. with the most stable location being at α = 1.25 AU and e « 0.2.," They find a broad, relatively stable region for test particles between 0.9 and 1.4 AU, with the most stable location being at $a$ = 1.25 AU and $e$ $<$ 0.2."1074 This is a region in. which Paper | found no stable test particle orbits., This is a region in which Paper 1 found no stable test particle orbits.1075 Fig., Fig.1076 8. shows a plateau in survivability between 1.0 and 1.25 AU., \ref{fig:74b} shows a plateau in survivability between 1.0 and 1.25 AU.1077 Dvorak (2003) found no trend in the results of their simulations of massive planets. and concluded that the presence of an unseen companion in the system was unlikely.," Dvorak (2003) found no trend in the results of their simulations of massive planets, and concluded that the presence of an unseen companion in the system was unlikely."1078 Further observations will shed light on this issue. although the survival rate of Saturns for the entire region with 0.9 AU (ας 1.2 AU. ex 0.15 suggests that this is a real possibility.," Further observations will shed light on this issue, although the survival rate of Saturns for the entire region with 0.9 AU $< a <$ 1.2 AU, $e \leq$ 0.15 suggests that this is a real possibility."1079 Note again that the best-fit orbit of the outer planet in this system has recently been revised to ag = 3.40 AU. e 2 0.58 (Naef 2004).," Note again that the best-fit orbit of the outer planet in this system has recently been revised to $a$ = 3.40 AU, $e$ = 0.58 (Naef 2004)."1080 The closer proximity and higher eccentricity of this planet strongly affects the dynamies between the two known plants., The closer proximity and higher eccentricity of this planet strongly affects the dynamics between the two known plants.1081 Both Dvorak (2003) and Paper [assume the orbital elements from Table | in their calculations., Both Dvorak (2003) and Paper I assume the orbital elements from Table 1 in their calculations.1082 We have found specific locations in four known extrasolar planetary systems in which Saturn-mass planets could exist on stable orbits., We have found specific locations in four known extrasolar planetary systems in which Saturn-mass planets could exist on stable orbits.1083 Such a planet would lie just below the detection threshold of current radial velocity surveys. and may be detected in the near future.," Such a planet would lie just below the detection threshold of current radial velocity surveys, and may be detected in the near future."1084 Table 3 summarizes our results. detailing the location 1n (4.0) space of each maximum in the survival rate for each of our four candidate systems.," Table 3 summarizes our results, detailing the location in $a,e$ ) space of each maximum in the survival rate for each of our four candidate systems."1085 If an additional planet is discovered in the stable region of one of these systems. it would mark the first successful prediction of a planet since John Couch Adams predicted the existence of Neptune in 1845 based on perturbations to Uranus? orbit.," If an additional planet is discovered in the stable region of one of these systems, it would mark the first successful prediction of a planet since John Couch Adams predicted the existence of Neptune in 1845 based on perturbations to Uranus' orbit."1086 Does the presence of a stable region imply the presence of a planet?, Does the presence of a stable region imply the presence of a planet?1087 Must all systems contain as many planets as they can?, Must all systems contain as many planets as they can?1088" Laskar (1996) speculated that ""a planetary system will always be in this state of marginal stability. as a result of its gravitational interactions.”"," Laskar (1996) speculated that “a planetary system will always be in this state of marginal stability, as a result of its gravitational interactions.”"1089" The ""packed planetary systems"" (PPS) hypothesis. presented in Paper 1 (see also Barnes Quinn. 2004). extends this idea by suggesting that all systems contain as many planets às they can dynamically. support without self-disrupting."," The “packed planetary systems” (PPS) hypothesis, presented in Paper 1 (see also Barnes Quinn, 2004), extends this idea by suggesting that all systems contain as many planets as they can dynamically support without self-disrupting."1090 Allsystems may be on the edge of stability. but observational constraints prevent the detection of smaller or more distant bodies which push apparently stable systems to this edge.," Allsystems may be on the edge of stability, but observational constraints prevent the detection of smaller or more distant bodies which push apparently stable systems to this edge."1091 The formation scenario of a planet of any size in between, The formation scenario of a planet of any size in between1092"claim that the circular velocity profiles of subhaloes are best fit by a parabolic function, which is shallower than NFW circular velocity profiles, and that this confirms their earlier result that the internal structure and kinematics of the Milky Way satellites are in good agreement with the subhaloes found in CDM simulations (Stoehretal.2002).","claim that the circular velocity profiles of subhaloes are best fit by a parabolic function, which is shallower than NFW circular velocity profiles, and that this confirms their earlier result that the internal structure and kinematics of the Milky Way satellites are in good agreement with the subhaloes found in CDM simulations \citep{Stoehr.etal:02}."1093. 'The persistence of the steep inner density structure in our satellite simulations conflicts with their assumed parabolic velocity profile., The persistence of the steep inner density structure in our satellite simulations conflicts with their assumed parabolic velocity profile.1094" Recent very high resolution simulations following the formation of CDM haloes in a cosmological context also have steep central density profiles in their subhaloes, which confirms our result (Diemandetal.2008;Springeletal. 2008).."," Recent very high resolution simulations following the formation of CDM haloes in a cosmological context also have steep central density profiles in their subhaloes, which confirms our result \citep{Diemand.etal:08,Springel.etal:08}."1095" To estimate satellite mass loss, researchers often use the impulse approximation."," To estimate satellite mass loss, researchers often use the impulse approximation."1096 The impulse approximation assumes that the perturbation time scale is much shorter, The impulse approximation assumes that the perturbation time scale is much shorter1097demonstrate a clear segregation between the cooling [ow and non-cooling How svstenis.,demonstrate a clear segregation between the cooling flow and non-cooling flow systems.1098 All of the cooling How clusters in our sample exhibit good agreement between their X-ray and strong-lensing masses., All of the cooling flow clusters in our sample exhibit good agreement between their X-ray and strong-lensing masses.1099 In. particular. for those clusters with measured arc redshifts. excellent agreement between the X-ray and strong lensing masses is observed.," In particular, for those clusters with measured arc redshifts, excellent agreement between the X-ray and strong lensing masses is observed."1100 For the non-cooling lows. the strong lensing masses exceed the X-ray masses by factors of 24.," For the non-cooling flows, the strong lensing masses exceed the X-ray masses by factors of $2-4$."1101 For the two intermediate svstems. where the detections of cooling Lows from the X-ray data are more marginal. the lensing masses are again enhanced with respect to the X-ray values. although by a smaller factor (1.7 2.0).," For the two intermediate systems, where the detections of cooling flows from the X-ray data are more marginal, the lensing masses are again enhanced with respect to the X-ray values, although by a smaller factor $1.7-2.0$ )."1102 We find excellent agreement5 with the results of Miralda Escucdé and Babul (1995) for the three (non-cooling How) clusters in common with that study: Abell 1689. 2163 and 2218.," We find excellent agreement with the results of Miralda Escudé and Babul (1995) for the three (non-cooling flow) clusters in common with that study; Abell 1689, 2163 and 2218."1103 Our results support the conclusions drawn by Allen (1996a.b) tiat thermal pressure dominates over magnetic pressure. turbiilence and bulk moions in the central regions oftve relaxed cooling-Iow clusters. and that the hvdrostatic assumption acopted in the X-ray analysis of such svstenis is valid.," Our results support the conclusions drawn by Allen (1996a,b) that thermal pressure dominates over magnetic pressure, turbulence and bulk motions in the central regions of the relaxed cooling-flow clusters, and that the hydrostatic assumption adopted in the X-ray analysis of such systems is valid."1104 In al cases where discrepancies between the X-rav and strong lensing masses occur. the clusters appear," In all cases where discrepancies between the X-ray and strong lensing masses occur, the clusters appear"1105consideration alone it is not possible to place a tightly constraining limit on the eccentricity of the svstem.,consideration alone it is not possible to place a tightly constraining limit on the eccentricity of the system.1106" The low luminosity state of 8.2 107 core + (2 - 10 keV). assuming a distance o[ 2.3 kpc. observed. with (Smithetal.2006). and the quiescence detections of a 3e upper limit at «1.0 LO""n core (2 - 10 keV). 6.0 107 core + (2 - 10 keV) and 4.1 107 core s 0.5 - 10 keV) by (Sakanoetal.2002).. //NIVE (Romanoetal.20009b) and //Epic-PPN. (DBozzoοἱal.2000). respectively. are not. consistent. with spherically. symmetric. smooth wind. Boneli-Llovle accretion (Bondi&Lovie 1944)."," The low luminosity state of 8.2 $\times$ $^{33}$ erg $^{-1}$ (2 - 10 keV), assuming a distance of 2.3 kpc, observed with \citep{SmithJ173912006} and the quiescence detections of a $\sigma$ upper limit at $<$ 7.0 $\times$ $^{32}$ erg $^{-1}$ (2 - 10 keV), 6.0 $\times$ $^{32}$ erg $^{-1}$ (2 - 10 keV) and 4.1 $\times$ $^{32}$ erg $^{-1}$ (0.5 - 10 keV) by \citep{Sakano2002}, /XRT \citep{Romano2009b} and /Epic-PN \citep{Bozzo2010XMM} respectively are not consistent with spherically symmetric, smooth wind Bondi-Hoyle accretion \citep{BondiHoyle1944}."1107. While the variations in emission could. result from NS separation changes around the orbit for some of the possible orbital configurations in this system. the lack of an uncerlving flux modulation in Fig.," While the variations in emission could result from NS separation changes around the orbit for some of the possible orbital configurations in this system, the lack of an underlying flux modulation in Fig."1108 5 and the inconsistency in significance of dillerent. outbursts at. approximately the same phase of the orbit shows that a smoothly varving stellar wind can not explain the outbursts observed., \ref{fig5} and the inconsistency in significance of different outbursts at approximately the same phase of the orbit shows that a smoothly varying stellar wind can not explain the outbursts observed.1109" Instead: we take this to show that the neutron star is in [act travelling through an inhomogeneous. non-svnunetrical stellar wind in the NPE τοῦ. 302 svstem. in accordance with the ~clumpy wind models"" of SEN'Ts 2007.. Negueruelaetal. 2008))."," Instead we take this to show that the neutron star is in fact travelling through an inhomogeneous, non-symmetrical stellar wind in the XTE $-$ 302 system, in accordance with the “clumpy wind models” of SFXTs \citealt{2007A&A...476..335W}, \citealt{2008AIPC.1010..252N}) )."1110 Looking at Lig., Looking at Fig.1111 2. i can be seen that the observation occurred. at the edge of one of the emission side peaks., \ref{fig2} it can be seen that the observation occurred at the edge of one of the emission side peaks.1112" Hence the higher level of emission seen in that data compared to the ‘quiescent data, could. be resulting rom observation during an orbital phase that shows a systematic increase in luminosity over the full LBIS/ISCAL iehteurve.", Hence the higher level of emission seen in that data compared to the `quiescent data' could be resulting from observation during an orbital phase that shows a systematic increase in luminosity over the full IBIS/ISGRI lightcurve.1113 Figure 5. shows the phase folded lighteurve of NTE 302 with all identified outbursts removed. we see that there is residual non-zero emission remaining in his lighteurve.," Figure \ref{fig5} shows the phase folded lightcurve of XTE $-$ 302 with all identified outbursts removed, we see that there is residual non-zero emission remaining in this lightcurve."1114 Taking the average count rates out of this data a luminosity of 2.9 1) eerg (18 - GOKkoV) is calculated. using the conversions outlined in Table 1.," Taking the average count rates out of this data a luminosity of 2.9 $\times$ $^{33}$ erg $^{-1}$ (18 - keV) is calculated, using the conversions outlined in Table 1."1115 The act that persistent. emission. is observed. in the full out-of-outburst. LBIS/ISCRI lishteurve. shows some underlying activity in the system.," The fact that persistent emission is observed in the full out-of-outburst IBIS/ISGRI lightcurve, shows some underlying activity in the system."1116 Following the arguments in the above xwagraph this activity is attributed to many small flares that can not be detected: individually but. sum. up to a detectable emission over the full length of the LBIS/ISGRI lighteurve., Following the arguments in the above paragraph this activity is attributed to many small flares that can not be detected individually but sum up to a detectable emission over the full length of the IBIS/ISGRI lightcurve.1117". Behaviour such as this was observed in the observations of Bozzoetal.(2010) and. supports the conclusions. of Romanoetal.(2009b)— that ""true quiescence is a rare state’.", Behaviour such as this was observed in the observations of \citet{Bozzo2010XMM} and supports the conclusions of \citet{Romano2009b} that `true quiescence is a rare state'.1118 This is a dillerent. behaviour to that observed in LOR 2619 (Clarketal.2009) where the removal of outbursts resulted in a relatively small change in the shape of the lighteurve., This is a different behaviour to that observed in IGR $-$ 2619 \citep{DaveJ17544} where the removal of outbursts resulted in a relatively small change in the shape of the lightcurve.1119 The presence of underlving periodic modulation of the emission in that svstem and the lack of it in this one is likely an orbital effect., The presence of underlying periodic modulation of the emission in that system and the lack of it in this one is likely an orbital effect.1120 The larger orbit of NPE 302. prohibits. observable modulation. being produced. by the orbital variations in cither a smooth stellar wind component or the frequency of low level Daring events., The larger orbit of XTE $-$ 302 prohibits observable modulation being produced by the orbital variations in either a smooth stellar wind component or the frequency of low level flaring events.1121 NTIZ 302 shows the extent of its transient. nature from the observation. where a deep quiescence was observed. followed. by rapid. [ares with dvnamic ranges of over 107. and took place at à. phase of 0.468 in this ephemeris. placing it within the periastron region.," XTE $-$ 302 shows the extent of its transient nature from the observation, where a deep quiescence was observed followed by rapid flares with dynamic ranges of over $^{3}$, and took place at a phase of 0.468 in this ephemeris, placing it within the periastron region."1122 ‘This suggests a very high degree of clumping” within the wind in order to allow quiescence states. in. very. close proximity to distinct. outbursts. to be observed. within the periastron region of the orbit.," This suggests a very high degree of `clumping' within the wind in order to allow quiescence states, in very close proximity to distinct outbursts, to be observed within the periastron region of the orbit."1123 This is taken to show that the majority of the periodicity information in NIS J1739-302 is a result of outburst emission., This is taken to show that the majority of the periodicity information in XTE J1739-302 is a result of outburst emission.1124 To safeguard against a smal number of events separated. by multiples of the proposec orbital period [alsely creating a signal. iterative and random removals of identified: outbursts from the lightcurve were performed.," To safeguard against a small number of events separated by multiples of the proposed orbital period falsely creating a signal, iterative and random removals of identified outbursts from the lightcurve were performed."1125 There were no indications from either case tha the removal of a small number of events could destroy the signal., There were no indications from either case that the removal of a small number of events could destroy the signal.1126 Ht is. believed. that the reliance. on outburs detection contributes to the periodicity only being seen by 1DIS/ISG Rasa combination of instrument sensitivities zu pointing strategies could prevent outburst detection in other instruments., It is believed that the reliance on outburst detection contributes to the periodicity only being seen by IBIS/ISGRI as a combination of instrument sensitivities and pointing strategies could prevent outburst detection in other instruments.1127 The presence of the two side-peaks seen in Fig. 2..," The presence of the two side-peaks seen in Fig. \ref{fig2},"1128 both detected at ~4 to 5e significance is intriguing and. could help in defining the orbital characteristics of the system., both detected at $\sim$ 4 to $\sigma$ significance is intriguing and could help in defining the orbital characteristics of the system.1129 In Ducciοἱal.(2009) an anisotropic stellar wind with an enhanced equatorial density region. inclined at some angle to the plane of the neutron star orbit. is invoked. within the clumpy wind. model to explain features in the phase-folded Lightcurve of LGR J112155952.," In \citet{Ducci2009} an anisotropic stellar wind with an enhanced equatorial density region, inclined at some angle to the plane of the neutron star orbit, is invoked within the clumpy wind model to explain features in the phase-folded lightcurve of IGR J11215–5952."1130 As a consequence a more general svstem was mocelled and it was shown that svnunetric outbursts corresponding to the crossing of the enhanced density region by the neutron star could. produce up to 3 outbursts per orbit (Figure 15 of Duccietal. 2009))., As a consequence a more general system was modelled and it was shown that symmetric outbursts corresponding to the crossing of the enhanced density region by the neutron star could produce up to 3 outbursts per orbit (Figure 15 of \citealt{Ducci2009}) ).1131" Using a simplified geometric version of this model with the equatorial density region inclined to the neutron star orbit at 90"". intersections of the neutron star orbit and stellar wind disc at the required phases are obtained for an orbit with an eccentricity of e 20.16."," Using a simplified geometric version of this model with the equatorial density region inclined to the neutron star orbit at $^{o}$, intersections of the neutron star orbit and stellar wind disc at the required phases are obtained for an orbit with an eccentricity of e $\sim$ 0.16."1132" However this results in a difference in separation between periastron and apastron of RI.. corresponding to à ratio of clump interaction probability of only Phere & PP, using the relationship in Clarketal. (2009)."," However this results in a difference in separation between periastron and apastron of $_{\odot}$, corresponding to a ratio of clump interaction probability of only $_{peri}$ $\simeq$ $_{ap}$ using the relationship in \citet{DaveJ17544}."1133". This is below the observed distribution of outbursts. Noo; = LSONN,,, for all outbursts or Ni; = NN, [or just the larger outburst events detailed in Table 1."," This is below the observed distribution of outbursts, $_{peri}$ = $_{ap}$ for all outbursts or $_{peri}$ = $_{ap}$ for just the larger outburst events detailed in Table 1."1134 This increase is to be expected in this model as clumps expauxd as they move out from the parent star. hence becoming Less dense and making strong outbursts less likely.," This increase is to be expected in this model as clumps expand as they move out from the parent star, hence becoming less dense and making strong outbursts less likely."1135 Since 90° is the most extreme case of the inclination of the enhanced, Since $^{0}$ is the most extreme case of the inclination of the enhanced1136with depth. whereas the magnetic field is two-dimensional.,"with depth, whereas the magnetic field is two-dimensional."1137 Fig., Fig.1138 Sbb shows the optical thickness of the tube., \ref{snapshot}b b shows the optical thickness of the tube.1139 Tere. the subphotospheric backeround is illustrated as an optically uck regiae and the atmosphere as an optically thin reenne.," Here, the subphotospheric background is illustrated as an optically thick regime and the atmosphere as an optically thin regime."1140 For better visibility the diameter of the ube iu Fig., For better visibility the diameter of the tube in Fig.1141 8 has been magnified by a factor 6.," \ref{snapshot}1142 has been magnified by a factor 6."1143 The arrows inside je tube indicate longitudinal flow velocities., The arrows inside the tube indicate longitudinal flow velocities.1144 Just below 1ο footpoiut a velocity of 3 kan + is present. bevoud 1| footpoiut the velocity increases to 6 kin 1. and at 16 outer part of the peuuubra near c=12 Maii the flow velocity reaches 13 km 1," Just below the footpoint a velocity of 3 km $^{-1}$ is present, beyond the footpoint the velocity increases to 6 km $^{-1}$, and at the outer part of the penumbra near $x=12$ Mm the flow velocity reaches 13 km $^{-1}$."1145 The footpoint of the flux tube can be identified with a bright pemmubral grain. because the upflow along the tube brings hot plasma to the surface which appears bright against the darker backeround.," The footpoint of the flux tube can be identified with a bright penumbral grain, because the upflow along the tube brings hot plasma to the surface which appears bright against the darker background."1146" As demoustrated iu the previous section. the tubes footpoiut migrates from the outer edge of the pemuubra msvard towards the παντα,"," As demonstrated in the previous section, the tube's footpoint migrates from the outer edge of the penumbra inward towards the umbra."1147 Iu reffootpoint the data points (plus sigux) are equally spaced iu time (At=300 3). illustrating that he footpoiut starts off at SO—13300 kan with a proper velocity exceeding 2 kin 1," In \\ref{footpoint} the data points (plus signs) are equally spaced in time $\triangle t=300$ s), illustrating that the footpoint starts off at $x=13\,300$ km with a proper velocity exceeding 2 km $^{-1}$."1148 After 12200 s (CIT data point from the right) the migration of the footpoiut decelerates from 1.5lm toxGl aus batt—7200 s These values are consistent with the proper motion of peuunibral eraius observed by Muller (19735)).," After 200 s $^{\rm th}$ data point from the right) the migration of the footpoint decelerates from 1.5 km $^{-1}$ to $\approx 0.1$ km $^{-1}$ at $t=7\,200$ s. These values are consistent with the proper motion of penumbral grains observed by Muller \cite{Mul73b}) )."1149 At the footpoint. the tubes temperature aud easons pressure are higher than the corresponding background values.," At the footpoint, the tube's temperature and gas pressure are higher than the corresponding background values."1150 As the plasima flows along. the tube loses internal energv by radiation.," As the plasma flows along, the tube loses internal energy by radiation."1151 Thereby it cools aud the gas pressure diminishes., Thereby it cools and the gas pressure diminishes.1152 At woxLO Mau it reaches temperature equilibrium with the surroundings., At $x \approx 10$ Mm it reaches temperature equilibrium with the surroundings.1153 As the gas pressure diminishes. the magnetic field streneth increases according to Eq. (13)).," As the gas pressure diminishes, the magnetic field strength increases according to Eq. \ref{pressure}) ),"1154 and flux conservation. Eq. (11)).," and flux conservation, Eq. \ref{flux_conservation}) ),"1155 προς that the tubes diameter decreases between the footpoiut and the point of thermal equilibriun., implies that the tube's diameter decreases between the footpoint and the point of thermal equilibrium.1156 Thus. our model is concurrent with observations frou Muller (L973a.. b) and Tritschler et al. (1997))," Thus, our model is concurrent with observations from Muller \cite{Mul73a}, b) and Tritschler et al. \cite{Tri97}) )"1157 which show that peuumubral exaius consist of a bright coma and a somewhat dimer thinner tail which is racially clongated and points away from the umbra., which show that penumbral grains consist of a bright coma and a somewhat dimmer thinner tail which is radially elongated and points away from the umbra.1158 ~Our mocel sugecsts that bright filameuts are the long tails «pemmubral erains., Our model suggests that bright filaments are the long tails of penumbral grains.1159 However. for lower spatial resolution. a few racially aligued peuuubral grains might iso be visible aud interpreted as oue bright filament.," However, for lower spatial resolution, a few radially aligned penumbral grains might also be visible and interpreted as one bright filament."1160 We want to mention. that the length of the bright tail depeucds ou the amount of magnetic flux ofthe tube: More magnetic flux nmuplies a larger diameter aud thus a lavecr optical depth. a longer radiative relaxation time (see Eq. (25))).," We want to mention, that the length of the bright tail depends on the amount of magnetic flux of the tube: More magnetic flux implies a larger diameter and thus a larger optical depth, a longer radiative relaxation time (see Eq. \ref{trad}) )),"1161 iid hence a louger bright tail., and hence a longer bright tail.1162 Iu view of Figs., In view of Figs.1163 Saa and sbh. one can see that the ube is optically thick as long as it is hotter than the surroundings.," \ref{snapshot}a a and \ref{snapshot}b b, one can see that the tube is optically thick as long as it is hotter than the surroundings."1164 This is due to the temperature dependence of the opacity which is the domunant opacity jr the photosphere: #&XT?., This is due to the temperature dependence of the $^-$ opacity which is the dominant opacity near the photosphere: $\tilde\kappa \propto T^{9}$.1165 Another important cature was already mentioned m Sect., Another important feature was already mentioned in Sect.1166 3.2.2: The tube is elevated. above the photosphere., 3.2.2: The tube is elevated above the photosphere.1167 That means that xieht filaments are optically thick structures overlying a darker backeround and surrounded by au optically thin atmosphere., That means that bright filaments are optically thick structures overlying a darker background and surrounded by an optically thin atmosphere.1168 Therefore. we propose that dark filaments might notexist s," Therefore, we propose that dark filaments might notexist ."1169"e, They rather appear as a consequence of spacing between racially cloneated bright filauneuts (Schlichemmaier et al. 1998)).", They rather appear as a consequence of spacing between radially elongated bright filaments (Schlichenmaier et al. \cite{Sli98}) ).1170 This is consistent with the statement of Muller (197205. 19735]) hat within he peuunibra bright features against a dark ckeround.," This is consistent with the statement of Muller \cite{Mul73a}, \cite{Mul73b}) ) that within the penumbra bright features against a dark background."1171 Our inodel also offers a consistent explanation for the Evershed effect: Between the footpoiut. at οzz10.5 Maa. éid the outer edge of the peuuubra. the matter iside the tube becomes gradually nore transparent.," Our model also offers a consistent explanation for the Evershed effect: Between the footpoint, at $x\approx 10.5$ Mm, and the outer edge of the penumbra, the matter inside the tube becomes gradually more transparent."1172 The optical thickness of the tube decreases to 7zm10+ (seo Fig., The optical thickness of the tube decreases to $\tau \approx 10^{-1}$ (see Fig.1173 Sbb} aud a line of sight tha crosses the tube reaches optical depth 7=2/3 iu the photosphere of the model at Do150 kn.id.e. below the tube.," \ref{snapshot}b b) and a line of sight that crosses the tube reaches optical depth $\tau=2/3$ in the photosphere of the model at $z=-150$ km, i.e. below the tube."1174 In other words. after the plasina has cooled reaching temperature equilibriun with the surroundings. the tube becomes transparent.," In other words, after the plasma has cooled reaching temperature equilibrium with the surroundings, the tube becomes transparent."1175 The tube constitutes a flow clhaunel that is elevated approximately 100 km above the photosphere and coufines an outward flow reaching velocities up to 13 lan +., The tube constitutes a flow channel that is elevated approximately 100 km above the photosphere and confines an outward flow reaching velocities up to $13$ km $^{-1}$ .1176 Inthe following, Inthe following1177mass derived from astrometric methods allows an independent estimate of the surface gravity g=0507/GMg or using the same fundamental parameters as employed for (2)).,mass derived from astrometric methods allows an independent estimate of the surface gravity $g=\varv_g^2c^2/GM_B$ or using the same fundamental parameters as employed for \ref{eqn:scaling}) ).1178 This leads to loge (mss77) =6.62+0.05., This leads to $\log g$ $^{-2}$ ) $= 6.62\pm 0.05$.1179 In our spectral modelling. we have discarded any solution where loge deviates more than 2c from the above value.," In our spectral modelling, we have discarded any solution where $\log g$ deviates more than $2\sigma$ from the above value."1180" For HZ 43A. ? gives a value of e,=30.1 kmss ? estimate the accuracy of this value to be 10-15 kmss'."," For HZ 43A, \citet{reid1996} gives a value of $\varv_g=30.1$ $^{-1}$ ; \citet{kruk2002} estimate the accuracy of this value to be 10–15 $^{-1}$."1181 Therefore we limit our. models to those cases where the gravitational redshift of HZ 43A is within the 10-50 ! range., Therefore we limit our models to those cases where the gravitational redshift of HZ 43A is within the 10–50 $^{-1}$ range.1182 We do this as follows., We do this as follows.1183 The parallax of HZ 43 A is known to be 15.3+2.9 mas (?).., The parallax of HZ 43 A is known to be $15.3\pm 2.9$ mas \citep{vanaltena1995}.1184 We use this to derive the minimum and maximum allowed distanced., We use this to derive the minimum and maximum allowed distance$d$.1185 For a given surface gravity. this distance range corresponds to an allowed range for gd when we use the accurately known value of fyRAE. Viilcderived from the optical flux (Sect. 2.3)).," For a given surface gravity, this distance range corresponds to an allowed range for $\varv_g=g d \sqrt{f_d} /c$ when we use the accurately known value of $f_d=R^2/d^2$ derived from the optical flux (Sect. \ref{sect:scaling}) )."1186 We then add this uncertainty in quadrature to à nominal uncertainty of 15 kmss7! in the observed redshift., We then add this uncertainty in quadrature to a nominal uncertainty of 15 $^{-1}$ in the observed redshift.1187 Solutions off by more than 2c are discarded., Solutions off by more than $2\sigma$ are discarded.1188" Also whenever the predicted range for v, does not overlap with the 10-50 ss7! range. we discard the solution."," Also whenever the predicted range for $\varv_g$ does not overlap with the 10–50 $^{-1}$ range, we discard the solution."1189 For HZ 43A. we already scale our spectra to agree with the optical flux of this star.," For HZ 43A, we already scale our spectra to agree with the optical flux of this star."1190 Our model spectra for Sirius B must also be in agreement with optical and UV flux measurements (Table 2p., Our model spectra for Sirius B must also be in agreement with optical and UV flux measurements (Table \ref{tab:siriusflux}) ).1191 We only consider here monochromatic fluxes. and avoid the use of magnitudes as this involves an extra complication. namely the convolution with filter transmissions.," We only consider here monochromatic fluxes, and avoid the use of magnitudes as this involves an extra complication, namely the convolution with filter transmissions."1192 Care should be taken in assessing the uncertainties in those flux measurements., Care should be taken in assessing the uncertainties in those flux measurements.1193 ? analysed the HST-GHRS Echelle-A spectrum of Sirius B. From their Fig., \citet{hebrard1999} analysed the HST-GHRS Echelle-A spectrum of Sirius B. From their Fig.1194 2 we measure the continuum flux at 1302 aas 1.270x1077 Wimm™ ss! AAT! or (8.32+0.07)x107 mm ss7! AA7!.," 2 we measure the continuum flux at 1302 as $1.270 \times119510^{-13}$ $^{-2}$ $^{-1}$ $^{-1}$ or $(8.32\pm 0.07) \times 10^4$ $^{-2}$ $^{-1}$ $^{-1}$."1196 Although the statistical errors are small. systematic effects are larger as shown below.," Although the statistical errors are small, systematic effects are larger as shown below."1197 ? have calibrated the GHRS Echelle-A spectrometer by comparing spectra with IUE spectra of the standard star jj Col., \citet{mack1997} have calibrated the GHRS Echelle-A spectrometer by comparing spectra with IUE spectra of the standard star $\mu$ Col.1198 From their Fig., From their Fig.1199 4c we find that the relative calibration accuracy of this instrument between 1250-1350 us 0.84+0.36 percent (the residual r.m.s., 4c we find that the relative calibration accuracy of this instrument between 1250–1350 is $\pm$ 0.36 percent (the residual r.m.s.1200 scatter between the GHRS and IUE spectra)., scatter between the GHRS and IUE spectra).

Showing the first 1,200 of 10274 lines. Download the file for the rest.