ReadingTimeMachine/rtm-sgt-ocr-v1

1source,target2 The simulations presented in 3.2 and 3.3 demonstrate that the bias of high-redshift galaxies can be measured directly from the variance in number counts between uncorrelated fields.," The simulations presented in \ref{subsection:uncorrelated_acs_fields} and \ref{subsection:uncorrelated_wfc3_fields}3 demonstrate that the bias of high-redshift galaxies can be measured directly from the variance in number counts between uncorrelated fields."4" Using the connection between bias and DM halo mass calculated from cosmological simulations (Tinkeretal.2008, 2010),, the galaxy bias can be used to estimate the characteristic mass of DM halos with similar spatial clustering strength."," Using the connection between bias and DM halo mass calculated from cosmological simulations \citep{tinker2008a,tinker2010a}, , the galaxy bias can be used to estimate the characteristic mass of DM halos with similar spatial clustering strength."5" For detailed discussions of the connection between halo mass and bias, we refer the reader to Robertson(2010) and Tinkeretal.(2010).."," For detailed discussions of the connection between halo mass and bias, we refer the reader to \cite{robertson2010a} and \cite{tinker2010a}."6 Figure 3 shows the characteristic halo mass estimated by converting the bias (Figure 1)) using the Tinkeretal. bias-mass relation determined from cosmological simulations.," Figure \ref{fig:mass_constraint}7 shows the characteristic halo mass estimated by converting the bias (Figure \ref{fig:bias_constraint}) ) using the \cite{tinker2010a} bias-mass relation determined from cosmological simulations."8" For the case of MuyS; —20.5AB magnitude i- dropout galaxies at redshift z~6 with bias b=7.5, the mass of halos with the same bias at z~6 is Mhalo72.1x10''h-'Mo (dashed line)."," For the case of $\MUV\lesssim-20.5$ AB magnitude $i$ -dropout galaxies at redshift $z\sim6$ with bias $b\approx7.5$, the mass of halos with the same bias at $z\sim6$ is $M_{\mathrm{halo}}\approx2.1\times10^{11} h^{-1}M_{\sun}$ (dashed line)."9" The bias can provide a precise estimate of a characteristic DM halo mass (8% in logM for Νῄειας=30 and 3% in logM for Νῃειας= 200) that, while skewed, is accurate (Ay=[(logM)—0.03 for Λειας=30 and Ay=—0.007 for —logM]/logMNgga,= 200)."," The bias can provide a precise estimate of a characteristic DM halo mass $8\%$ in $\log M$ for $\Nfields=30$ and $3\%$ in $\log M$ for $\Nfields=200$ ) that, while skewed, is accurate $\Delta_{M}\equiv[\ave{\log M}-\log M]/\log M = -0.03$ for $\Nfields=30$ and $\Delta_{M}= -0.007$ for $\Nfields=200$ )."10" This measure of a characteristic halo mass only corresponds to the halo mass of i-dropout galaxies if each halo hosts one galaxy and mass strongly correlates with luminosity (seeRobertson 2010),, but provides a convenient conceptual tool for discussing the approximate mass scale of high-redshift galaxies."," This measure of a characteristic halo mass only corresponds to the halo mass of $i$ -dropout galaxies if each halo hosts one galaxy and mass strongly correlates with luminosity \citep[see][]{robertson2010a}, but provides a convenient conceptual tool for discussing the approximate mass scale of high-redshift galaxies."11" We have presented a simple counts-in-cells method (e.g.,Peebles1980;Adelbergeretal.1998) for measuring the bias of high-redshift galaxies from the field-to-field variation in number counts induced by cosmic variance."," We have presented a simple counts-in-cells method \citep[e.g.,][]{peebles1980a,adelberger1998a} for measuring the bias of high-redshift galaxies from the field-to-field variation in number counts induced by cosmic variance."12" The number of high-redshift galaxies of a given luminosity are measured in a large number independent, widely-separated fields."," The number of high-redshift galaxies of a given luminosity are measured in a large number independent, widely-separated fields."13" The Poisson contribution to the variance in the number counts across these fields is removed, leaving a cosmic variance contribution that depends on the bias and themean matter overdensity fluctuations on the scale of the survey field volume."," The Poisson contribution to the variance in the number counts across these fields is removed, leaving a cosmic variance contribution that depends on the bias and themean matter overdensity fluctuations on the scale of the survey field volume."14 We use Monte Carlo simulations of, We use Monte Carlo simulations of15"The luminosity of scattered light per unit solid angle is given as: where N,=Mi, is the free electron number in the jet. µου the nucleon fiftoGram) electron number ratio. 77, the proton mass. and f4 the fraction of free tonized electrons.","The luminosity of scattered light per unit solid angle is given as: where $N_e = M_{\rm jet} f_{\rm el} / (\mu_e m_p )$ is the free electron number in the jet, $\mu_e$ the nucleon to electron number ratio, $m_p$ the proton mass, and $f_{\rm el}$ the fraction of free ionized electrons."16" Assuming that the et material is mostly heavy element.e.g.. C*O. we set ji,=2."," Assuming that the jet material is mostly heavy element,e.g., C+O, we set $\mu_e = 2$."17" Here. )/dO=(/Yozx)oy(1+cos? is the cross section of dott,Thomson scattering for unpolarized0,.) light."," Here, $d\sigma(\theta_{\rm obs}')/d\Omega = (3/16\pi) \sigma_T18(1 + \cos^2 \theta_{\rm obs}')$ is the cross section of Thomson scattering for unpolarized light."19 The scattered luminosity dL]./dO is related to that in. the supernova/observer restframe as (e.g.. Rybicki Lightman 1979): The ratio of polarized to unpolarized flux. fp. 1s given as: where II=(1—cos?0“4/1E04.)obs) is the degree of polarization of scattered wave. and the last factor of (12-2) is coming from the definition of fp by the ratio of flux per unit wavelength. f4.," The scattered luminosity $dL'_{\rm sc}/d\Omega$ is related to that in the supernova/observer restframe as (e.g., Rybicki Lightman 1979): The ratio of polarized to unpolarized flux, $f_P$, is given as: where $\Pi = (1 - \cos^2 \theta'_{\rm obs})/ (1 + \cos^2 \theta'_{\rm obs})$ is the degree of polarization of scattered wave, and the last factor of $(1+z)^{-1}$ is coming from the definition of $f_P$ by the ratio of flux per unit wavelength, $f_\lambda$."20 From eqs. (6)), From eqs. \ref{eq:t_sc}) )21" and (7)). the fractional time delay of scatterec photons direct unscattered photons Is (fap,—frau)/fot,=ΓΕ)from20.23. luminosity whichis small and independent of unknow1 Mop."," and \ref{eq:t_rad}) ), the fractional time delay of scattered photons from direct unscattered photons is $(t_{\rm obs} - t_{\rm rad})/ t_{\rm obs}22= 1 - (1+z)^{-1} = 0.23$ , which is small and independent of unknown $\theta_{\rm obs}$."23 Since the andspectrum of the supernova are not expected to change significantly within these time scales. we expect that the scattered spectrum is similar to the unscatterec one except for the redshift. as observed.," Since the luminosity and spectrum of the supernova are not expected to change significantly within these time scales, we expect that the scattered spectrum is similar to the unscattered one except for the redshift, as observed."24 Therefore we do not have to take into account the luminosity and spectral evolutior of supernova. Le.. fadfos.," Therefore we do not have to take into account the luminosity and spectral evolution of supernova, i.e., $t_{\rm rad} \sim t_{\rm obs}$."25 In the top and middle panels of Fig. L..," In the top and middle panels of Fig. \ref{fig:jet-energy},"26 we show J. 5. Mi. and Ejay=Miner—1) required to reproduce the observed redshift (z= 0.3) and degree of polarization (fp=1.8« 107). as a function of the jet viewing angle. (44.," we show $\beta$, $\gamma$, $M_{\rm jet}$ , and $E_{\rm jet} = M_{\rm jet} c^2 (\gamma -271)$ required to reproduce the observed redshift $z = 0.3$ ) and degree of polarization $f_P = 1.8 \times 10^{-3}$ ), as a function of the jet viewing angle, $\theta_{\rm obs}$."28" Here. we have assumed f;20.3 and£j,= 10d. and scaling of the results by different values of these parameters is obvious."," Here, we have assumed $f_{\rm el} = 0.3$ and $t_{\rm obs} = 10$ d, and scaling of the results by different values of these parameters is obvious."29 It is likely that there is another jet in. the opposite direction from the supernova. in addition to the jet considered so far. and the opposite jet should also produce another redshifted polarization component.," It is likely that there is another jet in the opposite direction from the supernova, in addition to the jet considered so far, and the opposite jet should also produce another redshifted polarization component."30 Since the data show only one redshifted component. the contribution from this opposite jet must be with similar polarization degree and redshift to the original jet. or negligibly small due to too small fp or very large redshift.," Since the data show only one redshifted component, the contribution from this opposite jet must be with similar polarization degree and redshift to the original jet, or negligibly small due to too small $f_P$ or very large redshift."31 In the bottom panel of Fig. 1..," In the bottom panel of Fig. \ref{fig:jet-energy},"32" we plot the redshift (5) and the polarization degree (fpop) by the opposite jet having the same jet velocity and mass but ns,=180°— Ou."," we plot the redshift $z_{\rm op}$ ) and the polarization degree $f_{P, \rm op}$ ) by the opposite jet having the same jet velocity and mass but $\theta_{\rm obs, op} = 180^\circ - \theta_{\rm33obs}$ ."34" This gives à constraint of (4,—100. otherwise we should have observed another polarized continuum component with larger fp and smaller redshift than the observed ones."," This gives a constraint of $\theta_{\rm obs} \lesssim 100^\circ$, otherwise we should have observed another polarized continuum component with larger $f_P$ and smaller redshift than the observed ones."35" The jet velocity is roughly constant at ./~0.2 for (s,2907. but it becomes more relativistic with decreasing Oop. at 90°. because the redshift effect of the jet motion is compensated by the blueshift to the observer."," The jet velocity is roughly constant at $\beta \sim 0.2$ for $\theta_{\rm36obs} \gtrsim 90^\circ$, but it becomes more relativistic with decreasing $\theta_{\rm obs}$ at $\theta_{\rm obs} \lesssim 90^\circ$ , because the redshift effect of the jet motion is compensated by the blueshift to the observer."37 The jet mass and energy rapidly becomes larger with decreasing (y. mainly because of less efficient polarization and larger rjxKe7fa.," The jet mass and energy rapidly becomes larger with decreasing $\theta_{\rm obs}$ , mainly because of less efficient polarization and larger $r_{\rm jet} \propto t_{\rm sc} \gg t_{\rm obs}$."38 Small OsC60) seems not favored from energetics. since it requires Jet energy of more than 10 erg.," Small $\theta_{\rm obs}39(\lesssim 60^\circ)$ seems not favored from energetics, since it requires jet energy of more than $10^{52}$ erg."40" These considerations lead to a conclusion that the jet directions must be close to (i,~907 (probably within + 10-207) and both the two jets contributed roughly equally to the observed polarization.", These considerations lead to a conclusion that the jet directions must be close to $\theta_{\rm obs} \sim 90^\circ$ (probably within $\pm$ $^\circ$ ) and both the two jets contributed roughly equally to the observed polarization.41" Therefore we assume two jets with (y,=907 in this work.", Therefore we assume two jets with $\theta_{\rm obs} = 90^\circ$ in this work.42 Then we found y=1—(14z)¢=0.23. and the jet mass is where Mj is re-defined as the mass of each jet.," Then we found $\beta = 1 -43(1+z)^{-1} = 0.23$, and the jet mass is where $M_{\rm jet}$ is re-defined as the mass of each jet."44 Therefore. observed redshifted polarization can be explained if the jet material is modestly ionized. with the kinetic jet energy Ej~5«10(MS/0.01M0.3/0.23 erg.," Therefore, observed redshifted polarization can be explained if the jet material is modestly ionized, with the kinetic jet energy $E_{\rm jet} \sim 5 \times4510^{50} (M_{\rm jet}/0.01 M_\odot) (\beta /0.23)^2$ erg."46" In the following of this paper we consistently use (y,=907. Ke=fa.) =0.23. and Mia70.01M..."," In the following of this paper we consistently use $\theta_{\rm obs} = 90^\circ$, $t_{\rm sc} =47t_{\rm obs}$, $\beta = 0.23$, and $M_{\rm jet} = 0.01 M_\odot$."48 First we consider the fact that the radio emission from SN 2002ap was very weak., First we consider the fact that the radio emission from SN 2002ap was very weak.49 If a considerable part of the jet kinetic energy was converted into nonthermal electrons via shock acceleration. inevitably there must be very strong radio emission which should have been even stronger than SN 1998bw.," If a considerable part of the jet kinetic energy was converted into nonthermal electrons via shock acceleration, inevitably there must be very strong radio emission which should have been even stronger than SN 1998bw."50 However. if the amount of CSM swept-up the jet is much smaller than the jet mass. the jet feels almost no deceleration and the majority of the jet material remains unshocked.," However, if the amount of CSM swept-up the jet is much smaller than the jet mass, the jet feels almost no deceleration and the majority of the jet material remains unshocked."51 We do not expect radio emission from such an almost freely expanding jet. and we only expect radio emission by CSM swept-up by the jet.," We do not expect radio emission from such an almost freely expanding jet, and we only expect radio emission by CSM swept-up by the jet."52 This emission and total energy of radio emitting electrons (~10? erg) are simply related by the jet velocity and CSM density. and related with the total jet mass and kinetic energy.," This emission and total energy of radio emitting electrons $\sim 10^{45}$ erg) are simply related by the jet velocity and CSM density, and related with the total jet mass and kinetic energy."53" Stellar wind mass-loss rate of Wolf-Rayet stars. which are considered as a possible candidate of the SN Ie progenitors. is typically M,~10°- 1072M ./yr with a wind velocity of V.—10? km/s (McCray 1983: Garefaa-Segura. Mae Low. Langer 1996; Garcíaa-Segura. Langer. Mac Low 1996)."," Stellar wind mass-loss rate of Wolf-Rayet stars, which are considered as a possible candidate of the SN Ic progenitors, is typically $\dot{M}_w54\sim 10^{-6}$ $10^{-5} M_\odot$ /yr with a wind velocity of $V_w \sim 10^3$ km/s (McCray 1983; Garcíaa-Segura, Mac Low, Langer 1996; Garc\'iaa-Segura, Langer, Mac Low 1996)."55 It is generally assumed that the CSM radio PLAN has a stellar wind profile. Le.. pesmi)around=Mif(br; ," It is generally assumed that the CSM around radio supernovae has a stellar wind profile, i.e., $\rho_{\rm CSM}(r) = \dot{M}_w /56(4 \pi r^2 V_w)$."57"Then the swept-up mass by the Jet becomes M.=Myra7.1.9«107bMyLsnltoM... where b=0.15.,Oi/4z is the beaming factor of the jet opening angle. {ο=fas 104. Ms=M,./U0°M.yr). and Va=V./0%km/s)."," Then the swept-up mass by the jet becomes $M_{\rm sw} = b58\dot{M}_w r_{\rm jet} / V_w = 1.9 \times 10^{-7} b_{-1} \dot{M}_{w, -6} V_{w,593}^{-1} t_{10} M_\odot$, where $b = 0.1 b_{-1} = \Omega_{\rm jet}/4\pi$ is the beaming factor of the jet opening angle, $t_{10} = t_{\rm obs}$ /10d, $\dot{M}_{w, -6} = \dot{M}_w / (10^{-6} M_\odot \rm /yr)$, and $V_{w, 3} =60V_w/(10^3 \rm km/s)$."61 This is much smaller than the jet mass and hence the jet is not decelerated., This is much smaller than the jet mass and hence the jet is not decelerated.62 Radio emission from SN 1998bw and 2002ap can be explained by isotropic high speed ejecta interacting with the CSM density consistent with the wind parameters. similar to the above values (Kulkarni et al., Radio emission from SN 1998bw and 2002ap can be explained by isotropic high speed ejecta interacting with the CSM density consistent with the wind parameters similar to the above values (Kulkarni et al.63 1998; Li Chevalier 1999: BKCO2). [, 1998; Li Chevalier 1999; BKC02). [64"The speed of ejecta is. however. considerably different for these two: 2,j~ a few for the former but (4~0.3c for the latter.]","The speed of ejecta is, however, considerably different for these two; $\gamma_{\rm ej} \sim$ a few for the former but $\upsilon_{\rm ej} \sim 0.3c$ for the latter.]"65 Since the inferred jet velocity of SN 2002ap ts close to (s.than and the swept-up CSM mass by the jet should be smaller that by isotropic ejecta because of the collimation. it looks reasonable that the radio emission from the external shock front of the jet is equal to or smaller than the observed radio emission.," Since the inferred jet velocity of SN 2002ap is close to $\upsilon_{\rm ej}$, and the swept-up CSM mass by the jet should be smaller than that by isotropic ejecta because of the collimation, it looks reasonable that the radio emission from the external shock front of the jet is equal to or smaller than the observed radio emission."66 We will present more detailed radio emission modeling in $4.., We will present more detailed radio emission modeling in \ref{section:radio}. .67 Therefore the weak radio flux from SN 2002ap does not immediately exclude the jet hypothesis. ifit is expanding freely.," Therefore the weak radio flux from SN 2002ap does not immediately exclude the jet hypothesis, ifit is expanding freely."68 It may also be useful torecall that the kinetic energy of supernovae(~10°! erg) is hardly converted into radiation. but supernovae are heated and shining by radioactivity.," It may also be useful torecall that the kinetic energy of supernovae$\sim 10^{51}$ erg) is hardly converted into radiation, but supernovae are heated and shining by radioactivity."69 However. free expansion raises another problem because of the expected rapid adiabatic cooling.," However, free expansion raises another problem because of the expected rapid adiabatic cooling."70 The jet material must be ionized at least, The jet material must be ionized at least71produced from this channel should have mass ratios AbifMoS 13. based on the ratio of allowed core A third. possibility is that. the core of the secondary star does not ignite helium at all.,"produced from this channel should have mass ratios $1.0 \leq72M_{1}/M_{2} \lessapprox 1.3$ , based on the ratio of allowed core A third possibility is that the core of the secondary star does not ignite helium at all."73 Phen the observed system would not contain two Le-burning stars. but one HLe-burning star and one post-RGB star that is simply a hot voung WD. cooling towards the main WD cooling sequence (11D 188112 seems to be an example of such à non-Le-burning hot subchwarl: see. e.g. Heber et 22003: Stroeer ct 22007).," Then the observed system would not contain two He-burning stars, but one He-burning star and one post-RGB star that is simply a hot young WD, cooling towards the main WD cooling sequence (HD 188112 seems to be an example of such a non-He-burning hot subdwarf; see, e.g., Heber et 2003; Stroeer et 2007)."74 The lifetime of this cooling phase is short (—1.10 Alvr. depending on its mass: Driebe et 11998: Heber et al.," The lifetime of this cooling phase is short $\sim 1-10$ Myr, depending on its mass; Driebe et 1998; Heber et al."75 2003)., 2003).76) Lt seems unlikely to us that the non-LHe-burning secondary. would. resemble the primary for long enough to reproduce the svstenis we are considering here. but. this possibility cannot be totally. excluded.," It seems unlikely to us that the non-He-burning secondary would resemble the primary for long enough to reproduce the systems we are considering here, but this possibility cannot be totally excluded."77 The couble-core CLE evolution. channel outlined: above is not the only possible way to make a binary containing two hot subcdwarfs., The double-core CE evolution channel outlined above is not the only possible way to make a binary containing two hot subdwarfs.78 Rappaport et ((2009) have outlined the evolutionary past and future of the spectroscopie binary ltegulus fa Leonis)., Rappaport et (2009) have outlined the evolutionary past and future of the spectroscopic binary Regulus $\alpha$ Leonis).79 The current low-mass (~ 0.39 M.) component could. potentially be a very low-mass πο star., The current low-mass $\sim$ 0.3 $\rm M_{\odot}$ ) component could potentially be a very low-mass sdB star.80 In one of the possible paths for the future evolution of the system. the core of Regulus (the current main-sequence star with mass  3.4 M. ) is exposed as a second sdB star. of mass ~ 0.5 M...," In one of the possible paths for the future evolution of the system, the core of Regulus (the current main-sequence star with mass $\sim$ 3.4 $\rm M_{\odot}$ ) is exposed as a second sdB star, of mass $\sim$ 0.5 $\rm M_{\odot}$."81 ln such a scenario. the lower-mass sc star could easily be still burning helium after the second one has been ormed.," In such a scenario, the lower-mass sdB star could easily be still burning helium after the second one has been formed."82 I£the current low-mass star did not manage to ignite iclium. then à system with only mareinally dillerent initial conclitions should be able to do so.," If the current low-mass star did not manage to ignite helium, then a system with only marginally different initial conditions should be able to do so."83 Phe Reeulus-like systems would produce a hot subdwarf binary with a mass ratio far rom unity (0.5/0.3zz 1.7). similar to the published mass ratio of PO. 1544|488.," The Regulus-like systems would produce a hot subdwarf binary with a mass ratio far from unity $\sim 0.5/0.3 \approx1.7$ ), similar to the published mass ratio of PG 1544+488."84 There may. well be even more channels which can ooduce double bot subdwarfs as binary evolution. allows or à rich range of possibilities., There may well be even more channels which can produce double hot subdwarfs as binary evolution allows for a rich range of possibilities.85 However. the double-core channel tends to. produce systems with mass ratios approaching one with minimal appeal to fine-tuning.," However, the double-core channel tends to produce systems with mass ratios approaching one with minimal appeal to fine-tuning."86 In addition. one distinguishing feature of both PC 1544|488 and LIE 0301-3039. is that they seem to be He-rich.," In addition, one distinguishing feature of both PG 1544+488 and HE 0301-3039 is that they seem to be He-rich."87. For a ltegulus-ike channel it is not so obvious why this should. produce abnormal sclB stars. as thev are simply a combination of normal formation channels.," For a Regulus-like channel it is not so obvious why this should produce abnormal sdB stars, as they are simply a combination of normal formation channels."88 Hence it seems reasonable to ask whether double-core evolution. might somehow tend to produce He-rich subdwarfs., Hence it seems reasonable to ask whether double-core evolution might somehow tend to produce He-rich subdwarfs.89 As explained above. the double-core channel strongly favours non-degenerate helium ignition and hence intermediate-mass stars (see Fie. 7)).," As explained above, the double-core channel strongly favours non-degenerate helium ignition and hence intermediate-mass stars (see Fig. \ref{fig:doublecore}) )."90 Figure SS shows that. during the relevant portion of their evolution. intermecliate-mass stars can have extended. regions outside their cores which have high. helium abuncanees. whilst low-mass stars do not.," Figure \ref{fig:compositionprofile}91 shows that, during the relevant portion of their evolution, intermediate-mass stars can have extended regions outside their cores which have high helium abundances, whilst low-mass stars do not."92 This seems to provide a natural reason why the known double subdwarf stars are Lf intermediate-mass stars tend to produce Le-rich hot παρναί». then the mass distribution. of that) subcwarl population should be less strongly peaked than the general sdB population.," This seems to provide a natural reason why the known double subdwarf stars are If intermediate-mass stars tend to produce He-rich hot subdwarfs, then the mass distribution of that subdwarf population should be less strongly peaked than the general sdB population."93" ""This is because those suchwarls from intermediate-nmass stars can ignite helium non-degenerately. unlike those low-mass stars which experience. the helium Uash."," This is because those sudwarfs from intermediate-mass stars can ignite helium non-degenerately, unlike those low-mass stars which experience the helium flash."94 We have calculated the evolution of post-sdB WD stars after they have acereted from (or merged. with) à helium WD companion., We have calculated the evolution of post-sdB WD stars after they have accreted from (or merged with) a helium WD companion.95 These stars burn helium in a shell around. the core. but spend a large part of their evolution with properties mainly determined by the sdD. mass.," These stars burn helium in a shell around the core, but spend a large part of their evolution with properties mainly determined by the sdB mass."96 Vheir later ractius evolution is mainly determined by the mass of helium gained from their WD companions., Their later radius evolution is mainly determined by the mass of helium gained from their WD companions.97 The only major uncertainty in our mocelling is in the treatment of the merger phase and the degenerate Ie ignition. but the merger calculations of Sato Jellery. (2000. 2002) seem to support our assumptions.," The only major uncertainty in our modelling is in the treatment of the merger phase and the degenerate He ignition, but the merger calculations of Saio Jeffery (2000, 2002) seem to support our assumptions."98 This demonstrates. that one. of the major. binary, This demonstrates that one of the major binary99slices.,slices.100 The significance of a detection is determined using mock catalogs containing he same number of objects as the data. as described in more detail in forthcoming vapers 0].," The significance of a detection is determined using mock catalogs containing the same number of objects as the data, as described in more detail in forthcoming papers ."101 Considering the 105 detections at the confidence level obtained from the color-slice analysis we show. in the left panel of Figure 1.. the relation between he (V.£) color characterizing the detected concentrations and the redshift.," Considering the 105 detections at the confidence level obtained from the color-slice analysis we show, in the left panel of Figure \ref{fig:vizmf}, the relation between the $(V-I)$ color characterizing the detected concentrations and the redshift."102 For he LIS candidates we take the redshift to be say as given by the matched-filter., For the EIS candidates we take the redshift to be $z_{MF}$ as given by the matched-filter.103 Lor zzi. we show the median and the quartiles of the color distribution of the detected red:7 galaxy concentrations., For $z\lsim0.7$ we show the median and the quartiles of the color distribution of the detected red galaxy concentrations.104 At higher redshifts. the error bars represent he full range of colors covered by the detections using deeper V data.," At higher redshifts, the error bars represent the full range of colors covered by the detections using deeper $V$ data."105 Also shown are curves corresponding to the expected variation of the (Vo£) color with redshift for a non-evolving elliptical galaxy ancl a passively evolving one. assuming a formation epoch at z24. a single burst of star formation. and a subsequent evolution in a low-density cosmological model.," Also shown are curves corresponding to the expected variation of the $(V-I$ ) color with redshift for a non-evolving elliptical galaxy and a passively evolving one, assuming a formation epoch at $z>4$, a single burst of star formation, and a subsequent evolution in a low-density cosmological model."106 The colors of carly-vpe galaxies along the red-sequence of spectroscopically confirmed. clusters are also shownο]., The colors of early-type galaxies along the red-sequence of spectroscopically confirmed clusters are also shown.107 From the figure we find that is bv and large a fair indicator of 1ο cluster redshift. albeit the large scatter. predicting colors consistent. with 1e models and empirical data.," From the figure we find that is by and large a fair indicator of the cluster redshift, albeit the large scatter, predicting colors consistent with the models and empirical data."108 The scatter in color is to a large extent due o the uncertainties in the detection procedure. errors in the galaxy colors and 10 uncertainties associated withz.," The scatter in color is to a large extent due to the uncertainties in the detection procedure, errors in the galaxy colors and the uncertainties associated with."109. Phe consistency between the mecians of re ELS candidate colors. the models and. the empirical data up τους <0.7 is remarkable.," The consistency between the medians of the EIS candidate colors, the models and the empirical data up to $z\lsim0.7$ is remarkable."110 For larger redshifts most cluster candidates are bluer than expected or their estimated redshift., For larger redshifts most cluster candidates are bluer than expected for their estimated redshift.111 From the visual inspection of these candidates. (in ilferent. passbands) we find that the blue colors are either due to contamination v a foreground cluster or they simply rellect the fact that galaxies in high-recshift clusters are not detected in the Y -band., From the visual inspection of these candidates (in different passbands) we find that the blue colors are either due to contamination by a foreground cluster or they simply reflect the fact that galaxies in high-redshift clusters are not detected in the $V$ -band.112 These results show that the elfectiveness of the (VW£2) color in confirming cluster candidates is in. practice limited. to systems with z<0.7. with deeper V. data providing little aditional leverage.," These results show that the effectiveness of the $(V-I)$ color in confirming cluster candidates is in practice limited to systems with $z\lsim0.7$, with deeper $V$ data providing little additional leverage."113 We havealso examined the CAl-cliagrams based on the available optical/infrared data for 15 clusters with zi20.6., We havealso examined the CM-diagrams based on the available optical/infrared data for 15 clusters with $z_{MF}\gsim0.6$.114 Prom the analysis of (/.—Avs) color slices. significant overdensities are detected near the nominal center of the candidates in 10 o£ the observed. fields.," From the analysis of $(I-Ks)$ color slices, significant overdensities are detected near the nominal center of the candidates in 10 of the observed fields."115 The same svstems are also identified in (/Avs). bu the weak dependence of this color on the redshift vields no independent constrain on the redshift.," The same systems are also identified in $(J-Ks)$, but the weak dependence of this color on the redshift yields no independent constraint on the redshift."116 For the 7954 confidence level detections. the colors range from about 2.4 t0 3.6 for (4.Avs) and 1.6 to 2.1 in (J.—Ns).," For the $>$ confidence level detections, the colors range from about 2.4 to 3.6 for $(I-Ks)$ and 1.6 to 2.1 in $(J-Ks)$."117 The dependence of the (£0dvs) color of cluster galaxies on redshift is shown in the right. panel of Figure 1.., The dependence of the $(I-Ks)$ color of cluster galaxies on redshift is shown in the right panel of Figure \ref{fig:vizmf}.118 For comparison we also show the expected dependence of the (£dys) color ofellipticals on redshift. for the sanie models presented earlier. and the relation for clusters observed speetroscopically18].," For comparison we also show the expected dependence of the $(I-Ks)$ color of ellipticals on redshift, for the same models presented earlier, and the color-redshift relation for clusters observed spectroscopically."119. Given the smal nuniber of candidates considered. we show the color of each detected: candidato.," Given the small number of candidates considered, we show the color of each detected candidate."120 The number in parenthesis indicates cases of overlap., The number in parenthesis indicates cases of overlap.121 While for low redshifts this color is a poor redshift indicator. it does give some leverage for redshifts 2 20.7.," While for low redshifts this color is a poor redshift indicator, it does give some leverage for redshifts $z\gsim0.7$ ."122 Note. in particular. that the colors of the ELS candidates are at least. consistent," Note, in particular, that the colors of the EIS candidates are at least consistent"123timescale and fy is the age of the cluster population (Cunedin&Os,timescale and $t_{H}$ is the age of the cluster population \citep{gne97}. .124trike, Fig.125r 1997).. Fie. d. shows that at least LGCs are above the thick line: this indicates that their preseut disruption timescales are <0.226j., \ref{fig:klessen} shows that at least $4$ GCs are above the thick line; this indicates that their present disruption timescales are $<0.22 t_{H}$.126 If the exponcut of distribution of lifetimes is 4~2. as derived for Galactic GCs (Cuedin&Ostriker 1997). the probability to have l out of 5 GCs with lifetimes less than 0.22£5; is ~1 whereas the probability that the lifetimes of all the GCs are less then θέ is O.L%.," If the exponent of distribution of lifetimes is $q\sim 2$ , as derived for Galactic GCs \citep{gne97}, , the probability to have $4$ out of $5$ GCs with lifetimes less than $0.22 t_{H}$ is $\sim 1\%$, whereas the probability that the lifetimes of all the GCs are less then $0.5 t_{H}$ is $0.4\%$."127 Hence the probability that the whole dark halo is comprised of objects with niasses >Πρι/0.2Mwpe7) 1M is less than1%.," Hence the probability that the whole dark halo is comprised of objects with masses $>5\times 10^{4} 128(\rho_{h}/0.02\,{\rm M}_{\odot}{\rm pc}^{-3})^{-1}$ $_{\odot}$ is less than$1\%$."129 If oulv a fraction f of the dark mass is iu compact objects of πας» Mj. then f<5«LOE NL/M.," If only a fraction $f$ of the dark mass is in compact objects of mass $M_{h}$, then $f<5\times 10^{4}$ $_{\odot}/M_{h}$."130 Localized regions with enhanced stellar density aud. where data pernüt. extremely cold kineniaties have been detected in some dSplis (e.g.. Olszewski&Aronson1985:Wlevnaetal. 2003.. hereafter N03: Colemanetal.2001:Walkeretal. 2006b3).," Localized regions with enhanced stellar density and, where data permit, extremely cold kinematics have been detected in some dSphs (e.g., \citealt{ols85, kle03}, hereafter K03; \citealt{col04,wal06b}) )."131 In particular. UAG dSph has received the most attention.," In particular, UMi dSph has received the most attention."132" Collecting the velocity of stars in 6' radius aperture. IK03 found that a two-Caussian »pulatious. oue representing the underblvius 8.8 kis Gaussian and the other with velocity dispersion o,=0.5 aufs. represcuting a subpopulation of fraction 0.7. is >3«LO! times more likely than the default 8.8 kus. The best-fit o, is iledetermined as it is much simaller han the median velocity errors (5 Καινή)."," Collecting the velocity of stars in $6'$ radius aperture, K03 found that a two-Gaussian populations, one representing the underlying $8.8$ km/s Gaussian and the other with velocity dispersion $\sigma_{s}=0.5$ km/s, representing a subpopulation of fraction $0.7$, is $>3\times 10^{4}$ times more likely than the default $8.8$ km/s. The best-fit $\sigma_{s}$ is ill-determined as it is much smaller than the median velocity errors $5$ km/s)."133 Nevertheless. even with these fiducial errors. we cau be certain that the velocity dispersion is «2.5 kms at ~95% confidence evel.," Nevertheless, even with these fiducial errors, we can be certain that the velocity dispersion is $<2.5$ km/s at $\sim 95\%$ confidence level."134 The stars which form the secondary density peals are not distinguished iu colour and maguitucde from the remauder of the CAG population Uslevnaetal.1998)., The stars which form the secondary density peak are not distinguished in colour and magnitude from the remainder of the UMi population \citep{kle98}.135. Tn act. UM star formation historv iucicates that its stars rave been formed in a suele burst earlier than 10 Car ago (Carreraetal.2002).," In fact, UMi star formation history indicates that its stars have been formed in a single burst earlier than $10$ Gyr ago \citep{car02}."136. Although UAG has long been suspected of experienciug ongoing tidal disruption. regious with euhanced volume density andkinematics camnot be the result of tidal interactions because the coarse-grained phase-deusity. ~po. iu collisionless svtoenis iust be constant or even decrease. thus implying that overdeusitv regions should appear dynamically hotter.," Although UMi has long been suspected of experiencing ongoing tidal disruption, regions with enhanced volume density and cannot be the result of tidal interactions because the coarse-grained phase-density, $\sim \rho/\sigma^{3}$, in collisionless sytems must be constant or even decrease, thus implying that overdensity regions should appear dynamically hotter."137 This suggests that the chuup is loung-ived., This suggests that the clump is long-lived.138 An alternative expliuation is that the density peak is a projection effect and that what we are secing is a cold. low-density tidal tail.," An alternative explanation is that the density peak is a projection effect and that what we are seeing is a cold, low-density tidal tail."139" However, umunuerical experiments have shown that this scenario is very uulikelv (Readetal.2006)."," However, numerical experiments have shown that this scenario is very unlikely \citep{rea06}."140.. The most plausible interpretation is that the clump is a disrupted stellar cluster. now surviving iu phase-space because the uuderling gravitational potential is harmonic (1x03).," The most plausible interpretation is that the clump is a disrupted stellar cluster, now surviving in phase-space because the underlying gravitational potential is harmonic (K03)."141 Within this potential. exavitational encouuters with the hypothetical VMOs will dominate the orbital diffusion of the stars once they become unbound from the progenitor cluster.," Within this potential, gravitational encounters with the hypothetical VMOs will dominate the orbital diffusion of the stars once they become unbound from the progenitor cluster."142 The integrity of the cold clamp may impose useful upper limits on the mass of VMOs., The integrity of the cold clump may impose useful upper limits on the mass of VMOs.143 The fact that the subpopulation is orbiting within the dark matter core of UA will ereatly simplify its dynamical description., The fact that the subpopulation is orbiting within the dark matter core of UMi will greatly simplify its dynamical description.144 Chimps stars will nudereo a random walk iun moment space by the collisious with the population of VMOs., Clump's stars will undergo a random walk in momentum space by the collisions with the population of VMOs.145 Tere we are interested iu the velocity change induced iu a star relative to the chuup ceuter of mass., Here we are interested in the velocity change induced in a star relative to the clump center of mass.146 The mean-square velocity change of a star in zu euncouuter with a VAIO of mass Mj. aud mipact parameter b255455. with ry) the chuup's median radius. in the impulsive approximation is: where V ds the masxinuun relative velocity hetween the chuup aud the perturber aud £2? is the meau-square position of the stars iu the clump (Spitzer1958).," The mean-square velocity change of a star in an encounter with a VMO of mass $M_{h}$, and impact parameter $b\geq 5 r_{\mathrm{\small 1/2}}$, with $r_{1/2}$ the clump's median radius, in the impulsive approximation is: where $V$ is the maximum relative velocity between the clump and the perturber and $\overline{r^{2}}$ is the mean-square position of the stars in the clump \citep{spi58}."147. In the opposite case of a head-on collision (b=0). the mean chauge is comparable to that predicted by the tidal approximation wheu b2L.ley75. ," In the opposite case of a head-on collision $b=0$ ), the mean change is comparable to that predicted by the tidal approximation when $b\simeq 1.4 r_{1/2}$."148The usual way to proceed is fo integrate Ac? eiven in Eq. (1)), The usual way to proceed is to integrate $\Delta \overline{v^{2}}$ given in Eq. \ref{eq:spitzer}) )149 for impact parameters b from Ἐνleyyy to ifnity and correct for the encounters in which the tidal approximation fails by a factor g23 (ee. Binney&Tremaine1987:Caclesetal. 2006)).," for impact parameters $b$ from $1.4 r_{1/2}$ to infinity and correct for the encounters in which the tidal approximation fails by a factor $g\approx 3$ (e.g., \citealt{bin87, gie06}) )."150" Doiug so. and for a distribution of Chuups and VALOs with a relative one-dimensional velocity dispersion 93,4. we obtain: For UNG. the persistence of the cbunip for a large fraction of a Uubble time indicates a core of the dark halo of at least 2-3 tines the size of the orbit of the chuup. which is 2150 pe."," Doing so, and for a distribution of clumps and VMOs with a relative one-dimensional velocity dispersion $\sigma_{\rm rel}$, we obtain: For UMi, the persistence of the clump for a large fraction of a Hubble time indicates a core of the dark halo of at least $2$ $3$ times the size of the orbit of the clump, which is $\gtrsim 150$ pc."151 In terms of the stellar core radius (~200 pe). this makes a halo core 1.5 2 times the stellar core and. consequently. the velocity dispersion of the halo particles in the core is. at least ~1.5 2 times the stellar velocity dispersion. corresponding to σι~15 20 kms. The impulsive approximation is valid for be<V. where w=Vrj anda. is the internal ouc-dinieusional velocity dispersion of the subpopulation.," In terms of the stellar core radius $\sim 200$ pc), this makes a halo core $1.5$ $2$ times the stellar core and, consequently, the velocity dispersion of the halo particles in the core is, at least $\sim 1.5$ $2$ times the stellar velocity dispersion, corresponding to $\sigma_{h}\sim 15$ $20$ km/s. The impulsive approximation is valid for $b\omega\leq V$, where $\omega=\sigma_{s}/r_{1/2}$ and $\sigma_{s}$ is the internal one-dimensional velocity dispersion of the subpopulation."152" For the encounters with bombr. responsible for most of the velocity impulse. this condition is well satisfied for a,Ὢνhoy."," For the encounters with $b\lesssim 5r_{1/2}$, responsible for most of the velocity impulse, this condition is well satisfied for $\sigma_{h}\gg 5 \sigma_{s}$."153" Therefore. for o,1l kin/s. this requirement is fulfilled within the isothermal dark core of UM."," Therefore, for $\sigma_{s}\sim 1$ km/s, this requirement is fulfilled within the isothermal dark core of UMi."154 Since stars in the clump are unbound. the selteravity of the chump in a first approximation cau be ignored considering oulv orbit diffusion iu the large-scale harmonic potential of the parent galaxx.," Since stars in the clump are unbound, the self-gravity of the clump in a first approximation can be ignored considering only orbit diffusion in the large-scale harmonic potential of the parent galaxy."155 In a one-dimensional harmonic potential a velocity impulse Ac? produces a change iu the velocity dispersion Ao?=A(c5Av?/2. where the brakets (..) refer to the ican value after averaging over one orbit.," In a one-dimensional harmonic potential, a velocity impulse $\Delta v^{2}$ produces a change in the velocity dispersion $\Delta \sigma^{2}\equiv \Delta \left<v^{2}\right>=\Delta v^{2}/2$, where the brakets $\left<...\right>$ refer to the mean value after averaging over one orbit."156 Combining this relation with Eq. (2)).," Combining this relation with Eq. \ref{eq:Deltav}) ),"157" we find the change of σε in a time Af: where og.)σι, is asstuued. since the population of chumps is expected to have a velocity dispersion similar to the stellar backeround. ~9 kin/s in UM."," we find the change of $\sigma_{s}$ in a time $\Delta t$: where $\sigma_{\rm rel}\approx \sigma_h$ is assumed, since the population of clumps is expected to have a velocity dispersion similar to the stellar background, $\sim 9$ km/s in UMi."158 The ratio Wefru dependsou the model: for a Phununer cluster y= L whereas g=1.5 for both a hing profile with a dimensionless potential depth of Hy=9 (c.e.. Cacleset 2006)) aud a Caussian density distribution.," The ratio $\eta\equiv \overline{r^{2}}/r_{1/2}^{2}$ dependson the model: for a Plummer cluster $\eta=4$ , whereas $\eta=1.5$ for both a King profile with a dimensionless potential depth of $W_{0}=9$ (e.g., \citealt{gie06}) ) and a Gaussian density distribution."159 Iu order to take a conservative value and to facilitate comparison with photometric and theoretical analvsis that assume Gaussian models. we will adopt jj=1.5 coustaut in time.," In order to take a conservative value and to facilitate comparison with photometric and theoretical analysis that assume Gaussian models, we will adopt $\eta=1.5$ constant in time."160 By fo; we will indicate the timerequired for a verycold group of unbound stars 04& 0. to acquire a velocity," By $t_{2.5}$ we will indicate the timerequired for a verycold group of unbound stars $\sigma_{s}\simeq 0$ , to acquire a velocity"161A far estimate of metal abuudauce in stellar systems still remains a central issue for any detailed. assessiueut of the other cluster distinctive propertics.,A fair estimate of metal abundance in stellar systems still remains a central issue for any detailed assessment of the other cluster distinctive properties.162 Metallicity modulates. iu fact. both effective temperature and apparent colors of stars. wlile the iuterual composition affects the nuclear cueine. aud therefore stellar lifetime.," Metallicity modulates, in fact, both effective temperature and apparent colors of stars, while the internal composition affects the nuclear engine, and therefore stellar lifetime."163" Even in case of resolved stellar svsteis. this cutaneled behavior may lead to a biased interpretation of cluster age. based on the CAID inorphologw. an effect often referred to as the ""ageauetallicitv. dilenuua (Reuzini&Duzzoni1986:Worthev 1991)."," Even in case of resolved stellar systems, this entangled behavior may lead to a biased interpretation of cluster age, based on the CMD morphology, an effect often referred to as the “age-metallicity dilemma” \citep{rb86,worthey94}."164. Such induced age uncertaiutv also reflects in the distance determination of star clusters. as far as one tries to compare the appare uaenitude of the Main Sequence Turn Off (TO) point with the appropriate theoretical hunuiuositv to derive herefrom the distauce modulus.," Such induced age uncertainty also reflects in the distance determination of star clusters, as far as one tries to compare the apparent magnitude of the Main Sequence Turn Off (TO) point with the appropriate theoretical luminosity to derive therefrom the distance modulus."165 One further difficulty also deals with the proper assessment of dust reddeuiug. rat may affect CAID morphology leading. iu geucral. to 1 artificially euliauced value of To overcome these problems. one would like xeliniuuilv derive the cluster metallicity from accura udauce analysis of individual stars (typically red its. due to thei brighter intrinsic hDnuninositv optical wavelengths) through high-resolution— spectroscopy (ee.Carretta&Crattou1997:raft&Ivans 2009).," One further difficulty also deals with the proper assessment of dust reddening, that may affect CMD morphology leading, in general, to an artificially enhanced value of To overcome these problems, one would like to preliminarily derive the cluster metallicity from accurate abundance analysis of individual stars (typically red giants, due to their brighter intrinsic luminosity at optical wavelengths) through high-resolution spectroscopy \citep[e.g.,][]{carretta97,kraft03}."166 This delicate task. however. is extremely time consunuüne. and a far more straight shortcut is often pursued relying on iutegrated cluster spectroscopy. usually taken at much lower resolution.," This delicate task, however, is extremely time consuming, and a far more straight shortcut is often pursued relying on integrated cluster spectroscopy, usually taken at much lower resolution."167" Narrow-buud spectroplotometric indices. as derived frou iutegrated low-res ( 6-8 FEWIIM) observations. usually provide the basic diagnostic scheme to match theoretical models frou stellar population svuthesis and derive therefrom cluster properties (ο,ο,,deFreitasPachecoetal.1998:Beasleyetal2002:Strader&Drodie 2001)."," Narrow-band spectrophotometric indices, as derived from integrated low-res $\sim$ 6-8 FWHM) observations, usually provide the basic diagnostic scheme to match theoretical models from stellar population synthesis and derive therefrom cluster properties \citep[e.g.,][]{pacheco98,beasley02,strader04}."168. The Lick svstem (Dwrsteiuotal.1981:Worthevct1991:Trageral.1998). stauds out as the widest aud most popular reference. assurng a systematic coverage of the main spectral features across the LLOQ64100 wwaveleneth range. casily accessible from erouud-based observations.," The Lick system \citep{burstein84,wortheyetal94,trager98} stands out as the widest and most popular reference, assuring a systematic coverage of the main spectral features across the 4100–6400 wavelength range, easily accessible from ground-based observations."169 Besides the advantage of this strateeyv. one cau still question whether it effectively allows us to ciscutanele any ageauctallicity degeneracy.," Besides the advantage of this strategy, one can still question whether it effectively allows us to disentangle any age-metallicity degeneracy."170 In fact. iutegrated iudices are sensitive both to iutriusic clemental abundance aud to the temperature distribution of the underlying stellar population. thus delivering a composite aud likely nou-univocal piece of iiformation (seeadiscussioniuDuzzoni1995b:Worthevetal.1995:Tautalo&Chiosi 2001).," In fact, integrated indices are sensitive both to intrinsic elemental abundance and to the temperature distribution of the underlying stellar population, thus delivering a composite and likely non-univocal piece of information \citep[see a discussion in][]{buzzoni95b,worthey95,171tantalo04}."172. An alternative. aud possibly more proficient wav out. at least for resolved stellar clusters. can be cuvisaged and will be explored in this paper.," An alternative, and possibly more proficient way out, at least for resolved stellar clusters, can be envisaged and will be explored in this paper."173 It relies ou a minimal, It relies on a minimal174analytically and with two dimensional magnetic wind solutions.,analytically and with two dimensional magnetic wind solutions.175" Although they are interested in T Tauri type stars, they claim that their models are independent of dynamo mechanism."," Although they are interested in T Tauri type stars, they claim that their models are independent of dynamo mechanism."176 Based on two-dimensional simulation they found that the field strength outside the star falls off as if with ne3.24 which is very similar to the simple dipolar field decay with n=3., Based on two-dimensional simulation they found that the field strength outside the star falls off as if with $n\approx 3.24$ which is very similar to the simple dipolar field decay with $n = 3$.177 We compare the values of the Alfvénn radius we calculate with those of Matt and jud-Doula&Owockij(2002) in Fig. ∙, We compare the values of the Alfvénn radius we calculate with those of \citet{matt2008a} and \citet{uddoula2002} in Fig. \ref{figmagconf}.178" In this comparison we used the dimensionless wind magnetic confinement parameter, 7 which characterizes the ratio of magnetic field energy density to the kinetic energy density of the wind."," In this comparison we used the dimensionless wind magnetic confinement parameter, $\eta$ which characterizes the ratio of magnetic field energy density to the kinetic energy density of the wind."179 We find a good agreement with that of ud-Doula&Owockij(2002)., We find a good agreement with that of \citet{uddoula2002}.180. However the Alfvénn radii calculated by us seem to be a factor of two smaller than those of |Matt&Pudritz|(2008a))., However the Alfvénn radii calculated by us seem to be a factor of two smaller than those of \citet{matt2008a}.181. It has been known for some time that many long-period Algol systems have accretion discs., It has been known for some time that many long-period Algol systems have accretion discs.182 Accreting material from such a disc should increase the spin rate of the more massive component up to its break-up speed as soon as even a small fraction of the mass has been transferred ∙∙, Accreting material from such a disc should increase the spin rate of the more massive component up to its break-up speed as soon as even a small fraction of the mass has been transferred \citep{demink2007}.183 All the classical Algols have a less massive evolved and a more massive main-sequence component., All the classical Algols have a less massive evolved and a more massive main-sequence component.184 Therefore a substantial amount of mass from the initially more massive star must be either lost or transferred to its companion., Therefore a substantial amount of mass from the initially more massive star must be either lost or transferred to its companion.185brown dwarf,brown dwarf.186 Both have hvdrogen-rich. envelopes., Both have hydrogen-rich envelopes.187 This provides a natural explanation Lor the low Ie abundance and surface gravity of (he remnant., This provides a natural explanation for the low He abundance and surface gravity of the remnant.188 The hydrogen is provided by the merged companion., The hydrogen is provided by the merged companion.189 Furthermore. (his companion also provides (he energy required (o eject (he envelope aud form the sdD. Several sdBs with low mass stellar ancl substellar companions have been [ound most recently and the true number may be much higher due to selection ellects 2011a.," Furthermore, this companion also provides the energy required to eject the envelope and form the sdB. Several sdBs with low mass stellar and substellar companions have been found most recently and the true number may be much higher due to selection effects ."190c).. A very important predicition made by is that sdBs formed via the CE-merger channel should be rare., A very important predicition made by is that sdBs formed via the CE-merger channel should be rare.191 222018—1916 is unique among 100 slowly rotating κα stars analvsed so far2009a)., $-$ 1916 is unique among $\simeq100$ slowly rotating sdB stars analysed so far.192. In contrast to that. predict a large fraction if not all of the single sdDs to be formed by WD-mergers.," In contrast to that, predict a large fraction if not all of the single sdBs to be formed by WD-mergers."193 Unless there is a mechanism to get rid off all Che angular momentum involved in a merger as suggested by(20021. tiis observation is hard (o explain.," Unless there is a mechanism to get rid off all the angular momentum involved in a merger as suggested by, this observation is hard to explain."194 Furthermore. predict that a laree fraction of the sdDs formed alter CE-merger should rotate with a critical velocity 044. which is defined as the rotational velocity al which mass loss induced by centrifugal forces prevents the red-giant core (o accrete more material [rom the secondary.," Furthermore, predict that a large fraction of the sdBs formed after CE-merger should rotate with a critical velocity $v_{\rm crit}$, which is defined as the rotational velocity at which mass loss induced by centrifugal forces prevents the red-giant core to accrete more material from the secondary."195" estimate this critical velocity to be about one third of the breakup velocity ei,=(GAL/R)Y?.", estimate this critical velocity to be about one third of the breakup velocity $v_{\rm br}=(GM/R)^{1/2}$.196 Using the parameters derived for 222018—1916. we caleulate coy7LSkms.| perfectly consistent with the projected rotational velocity measured [rom the spectrum.," Using the parameters derived for $-$ 1916, we calculate $v_{\rm crit}\simeq145\,{\rm km\,s^{-1}}$ perfectly consistent with the projected rotational velocity measured from the spectrum."197 In conclusion. the scenario proposed by and fits best with the observational data obtained so far. although the IIe-WD-4-IHe-WD or sdDB--IIe-WD merger scenarios cannot be ruled out.," In conclusion, the scenario proposed by and fits best with the observational data obtained so far, although the He-WD+He-WD or sdB+He-WD merger scenarios cannot be ruled out."198 222013— 1916 is the first candidate for a merger remnant among the hot subclwarl stars., $-$ 1916 is the first candidate for a merger remnant among the hot subdwarf stars.199 Similar objects are expected to be found in large spectroscopic databases like SDSS., Similar objects are expected to be found in large spectroscopic databases like SDSS.200 Due to the hieh rotational broadening the quality of these data should be sufficient to find them., Due to the high rotational broadening the quality of these data should be sufficient to find them.201 Based on observations at the La Silla Observatory of the European Southern Observatory or programmes number 032.D-0649 and 084.D-0348., Based on observations at the La Silla Observatory of the European Southern Observatory for programmes number 082.D-0649 and 084.D-0348.202 ο. G. is supported by the Deutsche Forschungsgemeinschaft (DEG) through grant. HE1356/49-1., S. G. is supported by the Deutsche Forschungsgemeinschaft (DFG) through grant HE1356/49-1.203 We thank L. \lorales-Ruecla or sharing her data with us., We thank L. Morales-Rueda for sharing her data with us.204 Furthermore. 5. C. wants to thank Ph.," Furthermore, S. G. wants to thank Ph."205 Podsiadlowski. C. S. Jelferv. Tt. IL. Ostensen and S.J. O'Toole for defending the merger channel as possible omnnation scenario for hot subcwarls.," Podsiadlowski, C. S. Jeffery, R. H. stensen and S. J. O'Toole for defending the merger channel as possible formation scenario for hot subdwarfs."206 Special thanks go to the organizers of the 4th sdOB neeling in Shanghai where (hese and other problems were discussed., Special thanks go to the organizers of the 4th sdOB meeting in Shanghai where these and other problems were discussed.207The absence of any obvious M-star feature im our spectruni can be used to derive a distance estimate to the syste.,The absence of any obvious M-star feature in our spectrum can be used to derive a distance estimate to the system.208 We assume that the iiaxiuun contribution of the ΑΙ:star at is and use as template spectrum for the secondary in RDBS0206 the AIG dwart 6106 with My=OMLOV W—7837.," We assume that the maximum contribution of the M-star at is and use as template spectrum for the secondary in RBS0206 the M6 dwarf Gl406 with $M_K = 9\fm19, V-K = 7\fm37$."209" This type of M-star would be appropriate for a cataclysinic binary with 2,4,90 atin.", This type of M-star would be appropriate for a cataclysmic binary with $P_{\rm orb} \sim 90$ min.210 The scaled V- and A-baud brightuesses of CLOG are Jl=2325 and A=1., The scaled $V$ - and $K$ -band brightnesses of Gl406 are $V_{\rm sc} = 23\fm5$ and $K_{\rm sc} = 16\fm1$.211" We assunee a Roche-lobe filling secondary star at a period of nuuin which has a spherical equivalent Roche radius of log(Ro/R:)= 0,86, ", We assume a Roche-lobe filling secondary star at a period of min which has a spherical equivalent Roche radius of $\log(R_2/R_{\sun}) = -0.86$ .212Usine Dailev's (1981) method combined with the imuiproved calibration of the surface brightuess of late-type stars by Beucrimanun Weichhold (1999.κ. --in prep.)," Using Bailey's (1981) method combined with the improved calibration of the surface brightness of late-type stars by Beuermann Weichhold (1999, in prep.)"213 which predicts a surface brightucss Sy=1.9 for a star like C106. the distance to RBS0206 is 7210 ppc.," which predicts a surface brightness $S_K = 4.9$ for a star like Gl406, the distance to RBS0206 is $> 240$ pc."214 Using the observed slope aud flux level of the white warf in the blue spectral regiae the «uecstions of white dwarf radius. feniperature a distance to the system can be addressed. too.," Using the observed slope and flux level of the white dwarf in the blue spectral regime the questions of white dwarf radius, temperature and distance to the system can be addressed, too."215 For hat exercise we use the model spectra for nonauagnetic white chart atinospheres by Caimusicke et al. (, For that exercise we use the model spectra for non-magnetic white dwarf atmospheres by Gännsicke et al. (216"1995) kiudlv provided bv D. Güuusicke,.",1995) kindly provided by B. Gännsicke.217 We assue a normal MAL: white wart with Rye=810 ccm.," We assume a normal $_{\sun}$ white dwarf with $R_{wd} = 8 \times 21810^8$ cm."219 The observed προςτα is reasonably well reflected with a IIS white dwarf at a distance of only ppc. although the model predicts a steeper spectral slope than observed.," The observed spectrum is reasonably well reflected with a K white dwarf at a distance of only pc, although the model predicts a steeper spectral slope than observed."220 A white dwarf at a disance of ppe as estimated above must have a considerably higher temperature of about 200001 in order to match the observe flux level atΑ., A white dwarf at a distance of pc as estimated above must have a considerably higher temperature of about K in order to match the observed flux level at.221. At this lieh teuperature the continu slope is much steeper han observed aud the fit in general is clearly worse than tha for TIS. Iu order to resolve the ciscreparcv between the different distance estimates one defiuitely needs yhase-resolved daa., At this high temperature the continuum slope is much steeper than observed and the fit in general is clearly worse than that for K. In order to resolve the discrepancy between the different distance estimates one definitely needs phase-resolved data.222 These would allow iu the first place to determine he orbita period., These would allow in the first place to determine the orbital period.223 Should our period estimate for sole reason be wrong aud the orbital period shorter fiai 90 iun. one can hide au even faiuter secondarv star with later spectral type iu the spectra which would be less dcistaut han the derived ppc for a period of nuin.," Should our period estimate for some reason be wrong and the orbital period shorter than 90 min, one can hide an even fainter secondary star with later spectral type in the spectrum which would be less distant than the derived pc for a period of min."224 Phase-resolved. spectroscopic data in the blue spectral reeinue would allow to diseutauele between radiation from the undisturbed plotosphere aud a ward accretion spot., Phase-resolved spectroscopic data in the blue spectral regime would allow to disentangle between radiation from the undisturbed photosphere and a warm accretion spot.225 The purity of the Zeeman spectrum shortward of Ila allows a direct measurement of he mean magnetic field streugth., The purity of the Zeeman spectrum shortward of $\alpha$ allows a direct measurement of the mean magnetic field strength.226 We fitted a second order polvnonuüal to interactively defined. continua points aud divided the observed spectrum by this curve., We fitted a second order polynomial to interactively defined continuum points and divided the observed spectrum by this curve.227 The resul is shown in Fie., The result is shown in Fig.228 6 together with a simple Zecman model dotted below the observed spectrum.," \ref{f:zeemod}229 together with a simple Zeeman model plotted below the observed spectrum."230 The model is based on the detailec coluputatious of the wavelengths aud oscillator streugtlis of the iudividial non-degenerate Zeeman transitious of U-Baluer lines bv Forster et al. (, The model is based on the detailed computations of the wavelengths and oscillator strengths of the individual non-degenerate Zeeman transitions of H-Balmer lines by Forster et al. (2311981) aud Rossuer et al. (,1984) and Rössner et al. (232198D).,1984).233 For he present purpose we use the trausitious of Ho. ITJ and I5. which lie iu the spectral range coverec bv our spectriii.," For the present purpose we use the transitions of $\alpha$, $\beta$ and $\gamma$, which lie in the spectral range covered by our spectrum."234 Our model just sus the oscillator strengths of al Bahuer transitions mentioned weightcc according to an assuned magnetic field distribution., Our model just sums the oscillator strengths of all Balmer transitions mentioned weighted according to an assumed magnetic field distribution.235 For the model show ain Fie., For the model shown in Fig.236 6 we assuned a Caussian fick distribution cered on B=36 MAIC with a spreac συ=2 NIMC., \ref{f:zeemod} we assumed a Gaussian field distribution centred on $B = 36$ MG with a spread $\sigma_B = 2$ MG.237 The model. therefore. does rot predict real spectral intensities but it predicts the waveleneths where spectral eatures are expected to occur.," The model, therefore, does not predict real spectral intensities but it predicts the wavelengths where spectral features are expected to occur."238 At the fiek strereth realized in RDS0206 nearly all subcomponeuts of the Balmer ines appear as individual nou-degeucerate rausitions due to the dominance of the quadratic over the lue:w Zeeman cfect., At the field strength realized in RBS0206 nearly all subcomponents of the Balmer lines appear as individual non-degenerate transitions due to the dominance of the quadratic over the linear Zeeman effect.239 Due to maguetic fiek slucaring and o the limited sxectral resolution these cannot be resolved in our spectrun. the Zeeman lines mainly appear as broad roughs.," Due to magnetic field smearing and to the limited spectral resolution these cannot be resolved in our spectrum, the Zeeman lines mainly appear as broad troughs."240 All features loneward of »belong to Πα. the features shortward ofthis wavelength »lougs to IL/ and II and become partly intermixed.," All features longward of belong to $\alpha$, the features shortward of this wavelength belong to $\beta$ and $\gamma$ and become partly intermixed."241 There are. however. some isolated features c.g. at aand rreacting sensitively on the adopted value of the ceutroid ficd streugth.," There are, however, some isolated features e.g. at and reacting sensitively on the adopted value of the centroid field strength."242 We estimate the uncertainty of the ceutroid field streneth to be about MMC., We estimate the uncertainty of the centroid field strength to be about MG.243 We reeard this field streugth as mcan plhotospheric field strenetl., We regard this field strength as mean photospheric field strength.244 Lf we want to, If we want to245Withwe more than 4109€ s o. exposure time. a siele Gaussian (e.g. Ixnóddlseder e al..,"With more than $10^9$ s of exposure time, a single Gaussian (e.g. Knöddlseder et al.,"246 2005) «Oes uot provide au acceptable des‘ription of the bee geometry., 2005) does not provide an acceptable description of the bulge geometry.247" We obtain a better adjustiuen of the data with two Ciaussiaus o[ FWHMNI 2.6 ? aud 1]LO * together with au extended component representing the Galactic clisx modelled by a Robi 1-3 Cur disk (+220° FWHM longitude extension. 5."" FWHM latitude exteusion)."," We obtain a better adjustment of the data with two Gaussians of FWHM 2.6 $^\circ$ and 11.0 $^\circ$, together with an extended component representing the Galactic disk modelled by a Robin 1-2 Gyr disk $\sim$ $^\circ$ FWHM longitude extension, $\sim$ $^\circ$ FWHM latitude extension)."248 This two Gasslal model has been used for siiplicity., This two Gaussian model has been used for simplicity.249 Aline one or lore sources simultaneotsly to the moclel. ike CIRS1758-258. GS1826-2| or H1713-22 kiJWLL to eiit above 100 keV. does no t[ect siguificantly te parameters of the Craussias: none of th sources are detected above 26.," Adding one or more sources simultaneously to the model, like GRS1758-258, GS1826-24 or H1743-22 known to emit above 100 keV, does not affect significantly the parameters of the Gaussians; none of these sources are detected above $2 \sigma$."250 However. a siguiica change is obtained by introσας ," However, a significant change is obtained by introducing 1E1740.7-2942."251"It may reflect simply tha ie bulge profile coitalis sInall-scale cdilfuse emission structu""es. localized in the central ~3°."," It may reflect simply that the bulge profile contains small-scale diffuse emission structures, localized in the central $\sim3^\circ$."252 Euthermore the modest aigular resolution of SPI makes it. difficult to distinguish between point sources. poiut-like and siiall-scale diffuse emission.," Furthermore the modest angular resolution of SPI makes it difficult to distinguish between point sources, point-like and small-scale diffuse emission."253 The Caussian parameters depend weakly ou he assumed disk. geomet‘y (for example he 2104 map is slightly peaked at the Galactic Center) but remain well iusle tle error bars., The Gaussian parameters depend weakly on the assumed disk geometry (for example the $240 \mu$ map is slightly peaked at the Galactic Center) but remain well inside the error bars.254 We also investigate a more complex bulge shape cosidering differen widths in / aud 6., We also investigate a more complex bulge shape considering different widths in $l$ and $b$.255 This gives Af=(LOl.1(1) aud Ab=(2.5nm1.0Y for the stralles Gaussian and Ad=(9.6-55) aud M3)° for the larger oue., This gives $\Delta l=(4.0_{-1.0}^{+1.4})$ and $\Delta b=(2.5_{-0.9}^{+1.0})^\circ$ for the smallest Gaussian and $\Delta l=(9.6_{-2.0}^{+2.5})$ and $\Delta b=(15.2_{-3.0}^{+3.0})^\circ$ for the larger one.256 The statistics do uot alow the derivation of aiy 1ueauingful conclusiou. but in all cases. these parameters stay compatible with tlose of the axisyumetric model.," The statistics do not allow the derivation of any meaningful conclusion, but in all cases, these parameters stay compatible with those of the axisymmetric model."257 In a following step. we let [ree he centroids of the Ciaussiais fy.by which are moving together.," In a following step, we let free the centroids of the Gaussians $l_0,~b_0$ which are moving together."258 The fit gives fy=(—0.61oye(15 all by=(0.060.190.20)'., The fit gives $l_0=(-0.64_{-0.19}^{+0.21})^{\circ}$ and $b_0=(0.06_{-0.20}^{+0.19})^{\circ}$.259 These vaues clepeud slightly on tje Choose configuratiou (in particular we tried several widths of the Gaussialls). iut all the paraimeters remalu compatible.," These values depend slightly on the choosen configuration (in particular we tried several widths of the Gaussians), but all the parameters remain compatible."260 Fig., Fig.261 10 shows the 1 o uucertaiuty zone iu tle longitude aud the latitute for tle centroid., \ref{fig:fig_contour} shows the 1 $\sigma$ uncertainty zone in the longitude and the latitute for the centroid.262 In this configuration the iut'oduction of an additional source is not 1eecded ανΠΙΟ6., In this configuration the introduction of an additional source is not needed anymore.263" To go further. we built a uode Cuposecl of a set of testec shells of €oustant densiy centered at (/.b)=(-0.6.0)"" with raclii 0-3.0. 3-7. 7-10. 10-15 and 15-197."," To go further, we built a model composed of a set of nested shells of constant density centered at $l,b$ $^\circ$ with radii 0-3.0, 3-7, 7-10, 10-15 and $^{\circ}$."264 We hus obtain the 51] keV flux clistributious just assuming :v racial eimiissiou., We thus obtain the 511 keV flux distributions just assuming a radial emission.265 We split each shell into heeative aud positive part [or the longitude aud then for t1e latitude., We split each shell into negative and positive part for the longitude and then for the latitude.266 Fie., Fig.267 LL and 12 display the racial profiles we obtained which appear quite colipatible with the coi1nation of the wo oll-centerecd axisvlninetrie Caussials determined above., \ref{fig:fig_radial_b} and \ref{fig:fig_radial_l} display the radial profiles we obtained which appear quite compatible with the combination of the two off-centered axisymmetric Gaussians determined above.268 We thus conclude lat tle centroid einission apyears slightly offset from Ιthe Galactic Centre cirection., We thus conclude that the centroid emission appears slightly offset from the Galactic Centre direction.269 This result. wule mareinally sieuilicant. is all the nore remarkable si1C Ixinzer et al. (," This result, while marginally significant, is all the more remarkable since Kinzer et al. ("270"2001) have found a siijlar sult of the bulge towards negative longitu 1 l OSSE/CGRO daa (lo shift measurecl )etween -0.25 ald -1"").",2001) have found a similar shift of the bulge towards negative longitude in the OSSE/CGRO data (1 $\sigma$ shift measured between -0.25 and $^{\circ}$ ).271" Finally. we cousicer that the best e«digration (hereafter called the reference moclel 1s] of 2 Gaussians centered at /jj=—0.67 axd bu=0.0"" representing the bulge plus au extenT"," Finally, we consider that the best configuration (hereafter called the reference model) consists of: 2 Gaussians centered at $l_0=-0.6^\circ$ and $b_0=0.0^\circ$ representing the bulge plus an extended disk"272with the bluest HB-type at a given [Fe/H] (see Figure 1)).,with the bluest HB-type at a given [Fe/H] (see Figure \ref{figure:GC_Groups}) ).273 In the study of the young halo GC population is defined as those clusters with a difference in in excess of —0.3 from that of the fiducial at a corresponding [Fe/H]., In the study of \citet{Mackey04} the young halo GC population is defined as those clusters with a difference in HB-type in excess of $-0.3$ from that of the fiducial at a corresponding [Fe/H].274 This corresponds to a minimum age difference of around -0.6 Gyr at HB-type — 0., This corresponds to a minimum age difference of around -0.6 Gyr at HB-type = 0.275 It should be noted that there is an important limitation the application of this technique for estimating the relative ages between clusters., It should be noted that there is an important limitation the application of this technique for estimating the relative ages between clusters.276 At the extremes of HB-type the isochrones become degenerate within observational uncertainties., At the extremes of HB-type the isochrones become degenerate within observational uncertainties.277 We note that parameters other than age can also be responsible for the location of a cluster in these groupings., We note that parameters other than age can also be responsible for the location of a cluster in these groupings.278" For example, in the case of NGC 2808 the extreme blue horizontal branch morphology is likely driven by He abundance variations."," For example, in the case of NGC 2808 the extreme blue horizontal branch morphology is likely driven by He abundance variations."279 In the study of Yoon&Lee(2002) relative age differences of around 1 Gyr are inferred from an examination of the mean period of type-ab RR Lyraes ((P.»)) with cluster metallicity., In the study of \citet{Yoon02} relative age differences of around 1 Gyr are inferred from an examination of the mean period of type-ab RR Lyraes $\langle P_{ab} \rangle$ ) with cluster metallicity.280 Oosterhoff Group I clusters possess (£55)~0.55 days and are more metal-rich than those of Group II with (£5)~0.65 days., Oosterhoff Group I clusters possess $\langle P_{ab} \rangle \sim 0.55$ days and are more metal-rich than those of Group II with $\langle P_{ab} \rangle \sim 0.65$ days.281" Yoon&Lee(2002) demonstrate that Group II clusters may be further split into ‘old’ and ‘young’ clusters (Groups II-a and II-b, respectively)."," \citet{Yoon02} demonstrate that Group II clusters may be further split into `old' and `young' clusters (Groups II-a and II-b, respectively)."282 The benefit of this method is that we can continue our classification to the blue extreme of HB-type., The benefit of this method is that we can continue our classification to the blue extreme of HB-type.283" A limitation on the method is that the (P,,) of many clusters is either limited by intrinsically low RR-ab numbers or the absence of studies in the literature 2009).", A limitation on the method is that the $\langle P_{ab} \rangle$ of many clusters is either limited by intrinsically low RR-ab numbers or the absence of studies in the literature \citep[see][]{Catelan09}.284. Our partitioning of the GGC population will rely on a combination of the latter two methods., Our partitioning of the GGC population will rely on a combination of the latter two methods.285" In the regime where both methods are viable (that is HB-type <0.8, and d) there are six objects in common: both methods classify five as ‘young’, the exception is the GC Rup 106 for which the ((P.,), [Fe/H]) method would suggest ‘old’ and the (HB-type, [Fe/H]) method would suggest *young'."," In the regime where both methods are viable (that is HB-type $< 0.8$, and $\langle P_{ab} \rangle > 0.60$ d) there are six objects in common: both methods classify five as `young', the exception is the GC Rup 106 for which the $\langle P_{ab} \rangle$, [Fe/H]) method would suggest `old' and the (HB-type, [Fe/H]) method would suggest `young'."286 Examination of the MSTO in the CMD demonstrates that it is indeed younger than most GCs (Marín-Franchetal.2009)., Examination of the MSTO in the CMD demonstrates that it is indeed younger than most GCs \citep{Marin-Franch09}.287". For the analysis that follows we will use the following criteria: Graphically, our partition of the GC population can be seen in Figure 1.."," For the analysis that follows we will use the following criteria: Graphically, our partition of the GC population can be seen in Figure \ref{figure:GC_Groups}."288" We take the [Fe/H], and in the analysis to follow, the distances and positions of GCs from the compilation of Harris(1996) (2010 edition)."," We take the [Fe/H], and in the analysis to follow, the distances and positions of GCs from the compilation of \citet{Harris96} (2010 edition)."289 The HBR is taken from the 2003 edition of the Harris catalog., The HBR is taken from the 2003 edition of the \citeauthor{Harris96} catalog.290 Koposov 1 2 are not included here as they lack sufficient dat to constrain their, Koposov 1 2 are not included here as they lack sufficient dat to constrain their291"rest frame they have a (vpical Lorentz factors,2»1 in a random direction.",rest frame they have a typical Lorentz factor $\gamma_e \gg 1$ in a random direction.292 We examine IC scattering of seed photons wilh a peak frequency tec and a peak fux £i (both measured al the observers rest frame)., We examine IC scattering of seed photons with a peak frequency $\nu_{\rm seed}$ and a peak flux $F_{\rm seed}$ (both measured at the observer's rest frame).293 We assume that the seed photons are roughly isotropic in the Πας frame., We assume that the seed photons are roughly isotropic in the fluid's frame.294 This would be the case if the seed photons are produced by svuchrotron radiation in the bulk. or any other mechanism local (o the moving fluid.," This would be the case if the seed photons are produced by synchrotron radiation in the bulk, or any other mechanism local to the moving fluid."295 We will consider External IC. in which the seed photons are produced by an external source elsewhere.," We will consider External IC, in which the seed photons are produced by an external source elsewhere."296 For simplicity we assume that all the photons have the same energy and all the electrons have the same Lorentz factor., For simplicity we assume that all the photons have the same energy and all the electrons have the same Lorentz factor.297 The energv and flux of the scattered photons are: and where Y=7476> and 7 are the Compton parameter ancl the optical depth in the Thomson scaltering regime., The energy and flux of the scattered photons are: and where $Y\equiv \tau \gamma_e^2$ and $\tau$ are the Compton parameter and the optical depth in the Thomson scattering regime.298 Note (hat (he unknown optical depth. 7. is introduced here in (he definition of Y but it is not used elsewhere in (he paper.," Note that the unknown optical depth, $\tau$, is introduced here in the definition of $Y$ but it is not used elsewhere in the paper."299 Our analvsis is independent of this unknown, Our analysis is independent of this unknown300perpendicular diffusion (üme-scale is shorter than the parallel scattering time-scale. i.e.. οι or in other terms the diffusion limit is (he dominant term in Eq. (33)),"perpendicular diffusion time-scale is shorter than the parallel scattering time-scale, i.e., $1/k_\parallel ^{min} v_\parallel$, or in other terms the diffusion limit is the dominant term in Eq. \ref{F1}) )"301" for large / and |<L/hey: D(a.z)ePla)— z""/a: therefore 1(2—q.gj)esin(qg/2)E(2q) (dashed ine in Fig.1))."," for large $t$ and $t < 1/k_\parallel ^{min} v_\parallel$: $\Gamma(a,z) \sim \Gamma(a) - z^a/a$ ; therefore $I(2-q, y_\parallel) \sim \sin(q\pi/2)\Gamma(2-q)$ (dashed line in \ref{drift}) )."302 In diffusive regine. the second term in Eq.(33)). representing the large scales dom2sn or hi«jum ). does not sienilicantlv contribute to the particle dift. as it is nanifest in Fig. l..," In diffusive regime, the second term in \ref{F1}) ), representing the large scales $l > 2\pi/k_\parallel^{min}$ or $k < k_\parallel^{min}$ ), does not significantly contribute to the particle drift, as it is manifest in Fig. \ref{drift}."303" We [ind (hat for slab turbulence. transverse particle drift coefficient rom local MEL is given by (hus subdifhisive with behaviour 87,0)~/?47 (depicted as the dashed line in Fig. 1))."," We find that for slab turbulence, transverse particle drift coefficient from local MFL is given by thus subdiffusive with behaviour $\kappa^s_D (t) \sim t^{-(2-q)}$ (depicted as the dashed line in Fig. \ref{drift}) )."304 Transverse subcliffusion has also been found by considering (me-scales longer than the parallel scattering (nme and (therefore allowing parallel scattering in Nota&Jokipii(2000)., Transverse subdiffusion has also been found by considering time-scales longer than the parallel scattering time and therefore allowing parallel scattering in \citet{kj00}.305. llowever. in that case particles are assumed {ο propagate back and forth along the MEL and to be Ged to the MEL.," However, in that case particles are assumed to propagate back and forth along the MFL and to be tied to the MFL."306 We notice that the drilt-coefficient me evolution in Eq. (35)), We notice that the drift-coefficient time evolution in Eq. \ref{slab_drift_asy}) )307 confirms that charged-particles in a turbulence depending on less (han 3 space coordinates remain confined within a gvroradius from the local field line (Jokipiietal.al. 1993," confirms that charged-particles in a turbulence depending on less than 3 space coordinates remain confined within a gyroradius from the local field line \citep{jkg93, jjb98}."308).The time-integration of Eq. (35)), .The time-integration of Eq. \ref{slab_drift_asy}) )309 up tof=Lj/vcPa[UenuQ). eives in case of weak turbulence (0B« By) the condition (Ar?)<ry.," up to $t = L_\parallel / v \sim 2\pi/(k^{min}_\parallel r_g \Omega)$, gives in case of weak turbulence $\delta B \ll B_0$ ) the condition $\langle \Delta x ^2 \rangle \ll r_g^2$."310 We notice that the time-integral of wy~P2 which provides ((Ar)?)e(81. ds an increasing function of ime for any observed physical value of ας however. as shown above. (his result does not contradict the (heoreni of reduced. dimensionalitv.," We notice that the time-integral of $\kappa^s_D \sim t^{q-2}$, which provides $\langle (\Delta x)^2 \rangle \sim t^{q-1}$, is an increasing function of time for any observed physical value of $q$; however, as shown above, this result does not contradict the theorem of reduced dimensionality."311 In sumuinary. (he present result has been obtained under three asstunplions: 1) ballistic motion in the 2 coordinate (2— (cj/): 2) average displacement transverse (o the local field B due to first-order drift: 3) Nolmogoroy power spectrum for mmagnelic fluctuations.," In summary, the present result has been obtained under three assumptions: 1) ballistic motion in the $z$ coordinate $z = v_\parallel t$ ); 2) average displacement transverse to the local field ${\bf B}$ due to first-order drift; 3) Kolmogorov power spectrum for magnetic fluctuations."312 Equations (32.. 33)) represent (hie average transverse displacement computed in the first-order orbit approximation at anv time smaller than the parallel scaltering ünme-scale. sothat the approximation of ballistic motion parallel to (he mean," Equations \ref{dXXlimit2}, \ref{F1}) ) represent the average transverse displacement computed in the first-order orbit approximation at any time smaller than the parallel scattering time-scale, sothat the approximation of ballistic motion parallel to the mean"313elliplicals. (heir study in galaxies within dillerent environments should help to discriminate between different formation aud evolution models.,"ellipticals, their study in galaxies within different environments should help to discriminate between different formation and evolution models."314 For instance. hierarchical scenarios predict that ellipticals in rich clusters. assembled completely al hieh redshilt (223). whereas field elliplicals may have experienced an elapsed and more complex star Formation history 1993)).," For instance, hierarchical scenarios predict that ellipticals in rich clusters assembled completely at high redshift $>$ 3), whereas field ellipticals may have experienced an elapsed and more complex star formation history \citealt{Kauff98}) )."315 However. verv little is known about the dependence of the relative abundances on environment.," However, very little is known about the dependence of the relative abundances on environment."316 One piece of information is (hat. there is no dilference in the |Meg/Fe] ratio between cluster and field elliptical galaxies (Jorgensen1999: al. 2002))., One piece of information is that there is no difference in the [Mg/Fe] ratio between cluster and field elliptical galaxies \citealt{Jor99}; \citealt{Kun02}) ).317 In this letter. we study Che behaviour of several Lick/IDS indices (see [or definition) in a sample of low and hieh density environment galaxies (LDEG and IIDEG respectively) and. surprisingly. we do find svstematic dillerences in the strength of C and CN features.," In this letter, we study the behaviour of several Lick/IDS indices (see \citealt{Worea94} for definition) in a sample of low and high density environment galaxies (LDEG and HDEG respectively) and, surprisingly, we do find systematic differences in the strength of C and CN features."318 Longslit spectra of 98 early(vpe galaxies in dillerent environments were taken in four observing runs with (wo different telescopes., Long–slit spectra of 98 early–type galaxies in different environments were taken in four observing runs with two different telescopes.319" The sample comprises 59 galaxies [rom the field and the Vireo cluster (LDEG). and 34 galaxies [rom the central region of the Coma cluster (IIDEG). spanning a wide range of absolute magnitudes (—22.5<Af, —16.5. using My75 km ! !). and central velocity dispersions 40<o400 kms | (from cwarf ellipticals. in Virgo ancl Coma. to eint galaxies)."," The sample comprises 59 galaxies from the field and the Virgo cluster (LDEG), and 34 galaxies from the central region of the Coma cluster (HDEG), spanning a wide range of absolute magnitudes $-22.5<M_{\rm B}<-16.5$ , using $H_0 =32075$ km $^{-1}$ $^{-1}$ ), and central velocity dispersions $40< \sigma <321400$ km $^{-1}$ (from dwarf ellipticals, in Virgo and Coma, to giant galaxies)."322 In the first (vo runs (1998 January and 1999 August) we used the 3.5m telescope at Calar Alto Observatory CAlmertaa. Spain). emploving the Twin Spectrograph.," In the first two runs (1998 January and 1999 August) we used the 3.5m telescope at Calar Alto Observatory a, Spain), employing the Twin Spectrograph."323 The observations of the third ancl fourth runs (1999 March and 2001 April) were carried out with the 4.2m WIHT al the Roque de los Muchachos Observatory (La Palma. Spain) using the ISIS spectrograph.," The observations of the third and fourth runs (1999 March and 2001 April) were carried out with the 4.2m WHT at the Roque de los Muchachos Observatory (La Palma, Spain) using the ISIS spectrograph."324 Spectral resolutions range from 2.6 aand 4.0 (FEWILIM) for LDEG to 8.6 [for ILDEG. in a spectral range AAZGOO 5400Α.," Spectral resolutions range from 2.6 and 4.0 (FWHM) for LDEG to 8.6 for HDEG, in a spectral range $\lambda\lambda 3600$ –5400."325. Exposures times of 12003600 secs per ealaxy allowed us to obtain central spectra wilh signal-to-noise (S/N) ratios (per À)) ranging from 25 to 250., Exposures times of 1200–3600 secs per galaxy allowed us to obtain central spectra with signal-to-noise $S/N$ ) ratios (per ) ranging from 25 to 250.326 We also observed several galaxies in common between runs (o ensure that (he measurements were in (he same svstem., We also observed several galaxies in common between runs to ensure that the measurements were in the same system.327 85 stars [rom the IDS/Lick library were included io transform the measured line-strength indices to the Lick svstem., $85$ stars from the IDS/Lick library were included to transform the measured line-strength indices to the Lick system.328 Standard data reduction procedures (fTlat-Delding. cosmic rav removal. wavelength calibration. skysubtraction and [Iuxing) were performed with (Cardiel 1999)). which allowed," Standard data reduction procedures (flat-fielding, cosmic ray removal, wavelength calibration, skysubtraction and fluxing) were performed with \citealt{Car99}) ), which allowed"329"Compared to the results by Ghisellinietal.(1998a),, we find temperatures that are significantly higher (up to a factor 3) at large optical depths.","Compared to the results by \citet{GHS98}, we find temperatures that are significantly higher (up to a factor 3) at large optical depths."330 This is probably due to a more precise treatment of the radiation field and Compton scattering., This is probably due to a more precise treatment of the radiation field and Compton scattering.331 Ghisellinietal.(1998a) considered only the cooling of particles by inverse Compton scattering and assumed that it was limited in the Thomson regime., \citet{GHS98} considered only the cooling of particles by inverse Compton scattering and assumed that it was limited in the Thomson regime.332" By using the exact Klein-Nishina cross section, we find more rapid photon escape and a weaker radiation field, whose cooling efficiency is lower."," By using the exact Klein-Nishina cross section, we find more rapid photon escape and a weaker radiation field, whose cooling efficiency is lower."333" At large Thomson optical depth (i.e. at high injection rates), Coulomb exchange is supposed to dominate over synchrotron self-absorption."," At large Thomson optical depth (i.e. at high injection rates), Coulomb exchange is supposed to dominate over synchrotron self-absorption."334" To investigate this, we completed the same simulations including e-e Coulomb scattering."," To investigate this, we completed the same simulations including e-e Coulomb scattering."335 Results are shown in Fig. 7.., Results are shown in Fig. \ref{GHS98D}.336 It is found that the e-e Coulomb collisions tend to increase the effective temperature., It is found that the e-e Coulomb collisions tend to increase the effective temperature.337" As explained before, particles are injected at high energy."," As explained before, particles are injected at high energy."338" They are cooled by both synchrotron emission and Compton scattering, and form a low energy thermal pool."," They are cooled by both synchrotron emission and Compton scattering, and form a low energy thermal pool."339 high energy particles are then scattered by thermal electrons with e-e Coulomb collisions., high energy particles are then scattered by thermal electrons with e-e Coulomb collisions.340" The cooling of the high energy distribution is very efficient but the thermal pool of cool electrons gain energy by this interaction, giving higher effective temperatures."," The cooling of the high energy distribution is very efficient but the thermal pool of cool electrons gain energy by this interaction, giving higher effective temperatures."341 This effect is negligible at low injection rates when the temperature is so high that the injection energy has a value that is almost in the bulk of the distribution and there is no well-marked high energy tail., This effect is negligible at low injection rates when the temperature is so high that the injection energy has a value that is almost in the bulk of the distribution and there is no well-marked high energy tail.342" However, at high injection rates, the temperature decreases and particles are injected at far higher energies than in the thermal pool."," However, at high injection rates, the temperature decreases and particles are injected at far higher energies than in the thermal pool."343 Exchange of energy between high and low energy particles becomes very efficient and it is found that this effect is significant (up to a factor of 2 for I.= 100)., Exchange of energy between high and low energy particles becomes very efficient and it is found that this effect is significant (up to a factor of 2 for $l_e=100$ ).344 A more detailed study of the synchrotron boiler mechanism and its application to X-ray binaries will be addressed in future work., A more detailed study of the synchrotron boiler mechanism and its application to X-ray binaries will be addressed in future work.345" As a second example, we investigate the effect of Fermi, second order acceleration."," As a second example, we investigate the effect of Fermi, second order acceleration."346" We consider a magnetised (Ip= 1), isolated plasma of size R=5x107 cm (typical of X-ray binary coronae), with no injection of seed photons."," We consider a magnetised $l_B=1$ ), isolated plasma of size $R=5\times10^7$ cm (typical of X-ray binary coronae), with no injection of seed photons."347 The soft photons are emitted by synchrotron radiation of high energy particles., The soft photons are emitted by synchrotron radiation of high energy particles.348 The acceleration is modelled by the second order Fermi process and no particle is injected into the plasma., The acceleration is modelled by the second order Fermi process and no particle is injected into the plasma.349 Particles are assumed to be trapped and the Thomson optical depth is set to be τε=1., Particles are assumed to be trapped and the Thomson optical depth is set to be $\tau_e=1$.350 Pair production/annihilation and Coulomb collisions are neglected to focus on the role of particle acceleration., Pair production/annihilation and Coulomb collisions are neglected to focus on the role of particle acceleration.351" After a transient phase that depends on the initial conditions, particles and photons reach a steady state that depends only on the acceleration properties."," After a transient phase that depends on the initial conditions, particles and photons reach a steady state that depends only on the acceleration properties."352 We first investigate the role of the acceleration efficiency and the threshold energy is assumed to be far lower than the bulk of particles., We first investigate the role of the acceleration efficiency and the threshold energy is assumed to be far lower than the bulk of particles.353 Figure 8 presents the steady particle distributions and spectra for various values of the acceleration efficiency., Figure \ref{acc1} presents the steady particle distributions and spectra for various values of the acceleration efficiency.354" In all cases, the distribution is similar to a Maxwell-Boltzmann distribution."," In all cases, the distribution is similar to a Maxwell-Boltzmann distribution."355" As found in previous calculations, the diffusion in the momentum space produces a quasi-thermal distribution (e.g.Katarzyfiskietal.2006b) and in this case, the thermalisation is also helped by the synchrotron boiler mechanism."," As found in previous calculations, the diffusion in the momentum space produces a quasi-thermal distribution \citep[e.g.][]{Katar06} and in this case, the thermalisation is also helped by the synchrotron boiler mechanism."356 The spectrum is the sum of the low energy synchrotron emission and a hard tail resulting from the multiple Compton scattering of these soft photons from the highest energy particles., The spectrum is the sum of the low energy synchrotron emission and a hard tail resulting from the multiple Compton scattering of these soft photons from the highest energy particles.357 As the acceleration efficiency increases the steady distribution widens and moves to higher energies., As the acceleration efficiency increases the steady distribution widens and moves to higher energies.358" As a consequence,"," As a consequence,"359"In order to establish division lines to separate SF, composite and AGN galaxies in the S2N2 diagram, we generated contour plots for each category of galaxies (see Fig.","In order to establish division lines to separate SF, composite and AGN galaxies in the S2N2 diagram, we generated contour plots for each category of galaxies (see Fig."360 3)., 3).361" As our sample of galaxies is larger for the SF and composite galaxies, we used contours enclosing ~90% for those galaxies."," As our sample of galaxies is larger for the SF and composite galaxies, we used contours enclosing $\sim$ $\%$ for those galaxies."362" However, as AGN galaxies are less numerous, we used contours enclosing ~75% for composite and AGNs."," However, as AGN galaxies are less numerous, we used contours enclosing $\sim$ $\%$ for composite and AGNs."363 The contour plots shown in Fig., The contour plots shown in Fig.364" 3a delimit two tangent parallel lines, generating with this criterium division lines defined by Eqs. ("," 3a delimit two tangent parallel lines, generating with this criterium division lines defined by Eqs. ("3651) and (2).,1) and (2).366" In order to define a division line between composite and AGN galaxies, we sampled the plot area with parallel lines of Eq. ("," In order to define a division line between composite and AGN galaxies, we sampled the plot area with parallel lines of Eq. ("367"1) in bins of 0.02 dex, generating in this way histograms for composite and AGN galaxies, where Eq.","1) in bins of 0.02 dex, generating in this way histograms for composite and AGN galaxies, where Eq."368 3 corresponds to the intersection of both (1) are separated by: 98.8% of galaxies above this line are SF galaxies and correspond to 88% of the SF sample. (, 3 corresponds to the intersection of both (i) are separated by: $\%$ of galaxies above this line are SF galaxies and correspond to $\%$ of the SF sample. (369ii) are divided by:,ii) are divided by:370In the absence of errors in the determination of the observable quantities. the last two equations can be solved for (he mass and radius of the neutron star.,"In the absence of errors in the determination of the observable quantities, the last two equations can be solved for the mass and radius of the neutron star."371 ILowever. because of the particular dependences of Fi and A on the neutron star mass and radius (see also Fie.," However, because of the particular dependences of $\ftd$ and $A$ on the neutron star mass and radius (see also Fig."372" 1 in Ozzel 2006). the loci of mass-radius points that correspond to each observable intersect, in general. ab two distinct positions."," 1 in Özzel 2006), the loci of mass-radius points that correspond to each observable intersect, in general, at two distinct positions."373 Moreover. the diverse nature of uncertainties associated to each of the observables requires a formal assessment of the propagation of errors. which we present here.," Moreover, the diverse nature of uncertainties associated to each of the observables requires a formal assessment of the propagation of errors, which we present here."374 We assign a probability distribution function to each of the observable quantities ancl denote them by PCD)d D. (Εαν. and. P(A)dA.," We assign a probability distribution function to each of the observable quantities and denote them by $P(D)dD$ , $P(\ftd)d\ftd$, and $P(A)dA$."375 Because the various measurements that lead to the determination of the three observables are independent of each other. the total probability density is simply given by the product Qur goal is to convert this probability density into one over the neutron-star mass. AL. and radius. H8.," Because the various measurements that lead to the determination of the three observables are independent of each other, the total probability density is simply given by the product Our goal is to convert this probability density into one over the neutron-star mass, $M$, and radius, $R$."376 We will achieve this by making a change of variables from (he pair (ο.A) to GU.2) and then by marginalizing over distance.," We will achieve this by making a change of variables from the pair $(\ftd,A)$ to $(M,R)$ and then by marginalizing over distance."377 Formallv. this implies that where (ο.A/V.2) is the Jacobian of the transformation.," Formally, this implies that where $J(\ftd,A/M,R)$ is the Jacobian of the transformation."378 Ht is important to emphasize here (hat. eiven a clistance D. not all pairs of the observables (Fjp.A) can be obtained with real values for the neutron-star mass and radius.," It is important to emphasize here that, given a distance D, not all pairs of the observables $(\ftd,A)$ can be obtained with real values for the neutron-star mass and radius."379 For (his reason. (he final distribution will not be normalized. even if the (ος distributions of equation (3)) are.," For this reason, the final distribution will not be normalized, even if the three distributions of equation \ref{eq:firstdistrib}) ) are."380 In addition. the factor 1/2 appears in equation (4)) because nearly all pairs of the observables (Ευ.4) correspond to two distinct pairs of CM.D).," In addition, the factor $1/2$ appears in equation \ref{eq:transform}) ) because nearly all pairs of the observables $(\ftd,A)$ correspond to two distinct pairs of $(M,D)$."381 There is only a region of the parameter space for which the pair of observables corresponds to a single pair of values for the mass and radius., There is only a region of the parameter space for which the pair of observables corresponds to a single pair of values for the mass and radius.382 However. this region has zero volume and. (therefore. will not contribute to the final probability distribution alter we marginalize over distance.," However, this region has zero volume and, therefore, will not contribute to the final probability distribution after we marginalize over distance."383 We can nowuse the above expressions to calculate (he Jacobian of the transformation, We can nowuse the above expressions to calculate the Jacobian of the transformation384E,.385"LE The Boussinesq approximation requires oy, where A and d¢ are respectively characteristic values for the wave number and perturbedvelocity."," The Boussinesq approximation requires v, where $k$ and $\delta v$ are respectively characteristic values for the wave number and perturbedvelocity."386 With &~Κ.δ. this requirement simplifies to l.," With $k\sim \bb{k}\bcdot \bb{b}$, this requirement simplifies to 1."387" The quantity on the left is δε, the inverse of the Revuolds number."," The quantity on the left is $1/Re$, the inverse of the Reynolds number."388 The Doussinesq limit is therefore generally appropriate For our problem in the limit of large Revnolds number., The Boussinesq limit is therefore generally appropriate for our problem in the limit of large Reynolds number.389 A finite resistivity will always be present. directly alfecting the dynamics because of its role in modilving the constraint of field reezine.," A finite resistivity will always be present, directly affecting the dynamics because of its role in modifying the constraint of field freezing."390 The question arises as to whether resistivity may generally be ignored at all wavenumbers if the viscous diffusivitv 7 much exceeds the resislive diffusivity ay. (, The question arises as to whether resistivity may generally be ignored at all wavenumbers if the viscous diffusivity $\nu$ much exceeds the resistive diffusivity $\eta_B$. (391Unlike the transformation of the viscous stress tensor. the parallel and (ransverse resistivities do not differ profoundly [Spitzer 1962]. ancl we shall ignore the distinction here.),"Unlike the transformation of the viscous stress tensor, the parallel and transverse resistivities do not differ profoundly [Spitzer 1962], and we shall ignore the distinction here.)"392 The presence of Ohnmic resistance alters equation (??)) to by = OUR. where A7 Dd.is (he magnitude. of. (he wavenumber.," The presence of Ohmic resistance alters equation \ref{bR}) ) to b_R = v_R, where $k^2$ is the magnitude of the wavenumber."393 In both of. our worked examples. & may be taken as fry.," In both of our worked examples, $k$ may be taken as $k_Z$."394 It is a straightforward exercise to rework the dispersion lormula using the above lor by., It is a straightforward exercise to rework the dispersion formula using the above for $b_R$ .395 One obtains for the case of axisvinmetric disturbances.," One obtains for the case of axisymmetric disturbances,"396(e.g.Ixaulfmannctal.2004).,\citep[e.g.][]{Kau04}.397.. Le could. be possible tha those LGBRD events with no detected hosts occured in very low surface brightness galaxies and hence. with low stellar masses.," It could be possible that those LGRB events with no detected hosts occured in very low surface brightness galaxies and hence, with low stellar masses."398 As mentioned before the cllects of dust coul also prevent the detection. of events in high metallicity. (ancl dustv) galaxies., As mentioned before the effects of dust could also prevent the detection of events in high metallicity (and dusty) galaxies.399 Lo our models these host galaxies exist. (ancl are part of the elobal galaxy population) bu the observability cut-oll has been determined by using the current. observed. stellar. mass distribution., In our models these host galaxies exist (and are part of the global galaxy population) but the observability cut-off has been determined by using the current observed stellar mass distribution.400 Fig., Fig.401 1. shows the distribution of stellar masses of observed host. galaxies constructed from the data of Savaglioetal.(2009).. together with those predicted by our scenarios.," \ref{mass} shows the distribution of stellar masses of observed host galaxies constructed from the data of \citet{Sav09}, together with those predicted by our scenarios."402 As it can be seen from Fig. 1..," As it can be seen from Fig. \ref{mass},"403 the scenario which best reproduces the observed. stellar mass. distribution is that with Ze=0.6Z. (scenario 11.9. see also Table 1).," the scenario which best reproduces the observed stellar mass distribution is that with $Z_{\rm C}=0.6\, {\rm Z_{\odot}}$ (scenario II.3, see also Table 1)."404 Lower metallicity thresholds predict lower observed stellar masses or the host galaxies. while no metallicity threshold (scenario ]) predicts larger ones.," Lower metallicity thresholds predict lower observed stellar masses for the host galaxies, while no metallicity threshold (scenario I) predicts larger ones."405" In the sample of Savaglio (2009).. we find that S5 per cent of the studied: galaxies rave stellar masses over the range z1077.ΜΑΙ, (seealsoCastroCoronetal. 2008)."," In the sample of \citet{Sav09}, we find that $85$ per cent of the studied galaxies have stellar masses over the range $\approx 10^{8.5-10.3} M_\odot$ \citep[see also][]{Cas08}."406. Phe probability of getting a host ealaxy within this stellar mass range in scenario 11.9 is SS »er cent. while for scenarios L. H1 and L2 it is TO. 2. and 43 »er cent. respectively.," The probability of getting a host galaxy within this stellar mass range in scenario II.3 is 88 per cent, while for scenarios I, II.1 and II.2 it is 70, 2, and 43 per cent, respectively."407 Phen we conclude that scenario 119 »edietions agree [αἱ] well with observations while others zil. and. therefore in the following sections. we will focus only on this scenario.," Then we conclude that scenario II.3 predictions agree fairly well with observations while others fail, and therefore in the following sections, we will focus only on this scenario."408 We stress the fact that its predictions include the elfects of host galaxies observabilitv. hence. a proper comparison with observations can be made.," We stress the fact that its predictions include the effects of host galaxies observability, hence, a proper comparison with observations can be made."409 La order to contribute to the understanding of the nature of LCIUD host galaxies. we will also compare these predictions with the properties (not weighted bv host galaxy. observabilitv) of both the sample ofall galaxies with mean cold gas metallicities below 0.6Z. (hereafter low metallicity sample) and the complete galaxy population of the catalogue of DeLucia&Blaizot(2007)..," In order to contribute to the understanding of the nature of LGRB host galaxies, we will also compare these predictions with the properties (not weighted by host galaxy observability) of both the sample of all galaxies with mean cold gas metallicities below $0.6\, {\rm Z_{\odot}}$ (hereafter low metallicity sample) and the complete galaxy population of the catalogue of \citet{DeL07}."410 As a first step towards understanding the nature of host ealaxies. we analyse their stellar masses as a function of redshift.," As a first step towards understanding the nature of host galaxies, we analyse their stellar masses as a function of redshift."411 As shown in Fig. 2.," As shown in Fig. \ref{massred},"412 the mean stellar. mass of the host galaxies as a function of redshift’ predicted: by scenario 9 reproduces the observed mean trend quite well., the mean stellar mass of the host galaxies as a function of redshift predicted by scenario II.3 reproduces the observed mean trend quite well.413 From this figure. we can also see that host. galaxies are. on average. more massive than galaxies in the complete galaxy population while the latter are more massive than those in the low metallicity sample.," From this figure, we can also see that host galaxies are, on average, more massive than galaxies in the complete galaxy population while the latter are more massive than those in the low metallicity sample."414 This can be understood taking into account that the host galaxy observabilitv is a strong function of the star formation rate.and that the complete galaxy. population in the catalogue of DeLuciaBlaizot(2007) follows à mass-mectallicity relationship (c.g.DeRossietal. 2009).," This can be understood taking into account that the host galaxy observability is a strong function of the star formation rate,and that the complete galaxy population in the catalogue of \citet{DeL07} follows a mass-metallicity relationship \citep[e.g.][]{DeR09}."415. Then. the cut-olf adopted. for the mean cold gas metallicity to reproduce the observed. stellar mass distribution implies a eut-olf in stellar mass since low metallicity galaxies are. on average. less massive than the eeneral galaxy population (see also Fig. 1)).," Then, the cut-off adopted for the mean cold gas metallicity to reproduce the observed stellar mass distribution implies a cut-off in stellar mass since low metallicity galaxies are, on average, less massive than the general galaxy population (see also Fig. \ref{mass}) )."416 However. as the observabilitv of a host galaxy depends strongly on its star formation activity and most of the small galaxies have low star formation rates. the observable host galaxies tend to be. on average. the more massive ones among them.," However, as the observability of a host galaxy depends strongly on its star formation activity and most of the small galaxies have low star formation rates, the observable host galaxies tend to be, on average, the more massive ones among them."417 As a result. our observable sample tends to be populated. by. svstenis more massive than those in the low metallicity sample or in the complete galaxy. population.," As a result, our observable sample tends to be populated by systems more massive than those in the low metallicity sample or in the complete galaxy population."418 In Fig. 3..," In Fig. \ref{gsfr},"419 we displav the mean SER. of host. galaxies as a function of redshift’ predicted: by scenario 1.9. together with the corresponding mean values for the low metallicity sample and the complete galaxy. population.," we display the mean SFR of host galaxies as a function of redshift predicted by scenario II.3, together with the corresponding mean values for the low metallicity sample and the complete galaxy population."420 As it can be seen. the prediction of scenario H.3 reproduces very well the behaviour of the observed host. galaxies.," As it can be seen, the prediction of scenario II.3 reproduces very well the behaviour of the observed host galaxies."421 Phese have higher SEI than the mean of the complete galaxy population. anc much higher than that of the low metallicity sample.," These have higher SFR than the mean of the complete galaxy population, and much higher than that of the low metallicity sample."422" The &ood agreement between our scenario L3 and observations suggests that the observed host galaxies are biased. towards galaxies. with. stellar masses in: the range 10""' INI... high", The good agreement between our scenario II.3 and observations suggests that the observed host galaxies are biased towards galaxies with stellar masses in the range $10^{9-10} {\rm M_\odot}$ high423" The &ood agreement between our scenario L3 and observations suggests that the observed host galaxies are biased. towards galaxies. with. stellar masses in: the range 10""' INI... high.", The good agreement between our scenario II.3 and observations suggests that the observed host galaxies are biased towards galaxies with stellar masses in the range $10^{9-10} {\rm M_\odot}$ high424SDSS Imaging Finding Chart Tool?.. and we identify galaxies that are clearly undergoing a merger event.,"SDSS Imaging Finding Chart Tool, and we identify galaxies that are clearly undergoing a merger event."425 We tind no enhancement in the merger fraction for the galaxies in the HI sample (9. 8. and 7 galaxies out of a total of 519 in the HI. Cj. and Ομ samples exhibit clear signs of a merger or interaction in our images).," We find no enhancement in the merger fraction for the galaxies in the HI sample (9, 8, and 7 galaxies out of a total of 519 in the HI, $_{M*}$, and $_{M*,nuvr}$ samples exhibit clear signs of a merger or interaction in our images)."426 As shown in Figures 2 and 3. galaxies from the HI sample are bluer. and have larger sizes and later type morphologies compared to Cy. galaxies.," As shown in Figures 2 and 3, galaxies from the HI sample are bluer, and have larger sizes and later type morphologies compared to $_{M*}$ galaxies."427 This is not surprising. because it is well known that galaxies that are more actively star-forming also contain more gas.," This is not surprising, because it is well known that galaxies that are more actively star-forming also contain more gas."428 The Cii;si; and Cyrnye galaxies. on the other hand. are matched to the HI sample both in stellar mass and in global NUV-r colour. so if HI gas fraction and star formation activity track euch other very closely. one might expect the galaxies in these two samples to have identical properties.," The $_{M*,NUV-r}$ and $_{M*,NUV-r,\mu*}$ galaxies, on the other hand, are matched to the HI sample both in stellar mass and in global $r$ colour, so if HI gas fraction and star formation activity track each other very closely, one might expect the galaxies in these two samples to have identical properties."429 In the following sections. we will show that this is not true.," In the following sections, we will show that this is not true."430 We divide the HI sample into four stellar mass bins. and study how sizes and colour gradients vary with atomic gas mass fraction.," We divide the HI sample into four stellar mass bins, and study how sizes and colour gradients vary with atomic gas mass fraction."431 We also compare properties of the corresponding control galaxies along the same sequence., We also compare properties of the corresponding control galaxies along the same sequence.432 The main purpose of comparing our results with those derived from control samples is to isolate those trends that can be attributed to increasing HI content. rather than to any other correlated property. such as stellar mass or global star formation rate.," The main purpose of comparing our results with those derived from control samples is to isolate those trends that can be attributed to increasing HI content, rather than to any other correlated property, such as stellar mass or global star formation rate."433 Because the samples have all been matched in redshift. the noise in the measurements of luminosity and colour necessary to estimate quantities such as stellar mass. will be identical in all the comparison samples.," Because the samples have all been matched in redshift, the noise in the measurements of luminosity and colour necessary to estimate quantities such as stellar mass, will be identical in all the comparison samples."434 Figures 3 and 4+ demonstrate that the galaxies in theHI sample have larger average sizes than both Cy. and Carr; control sample galaxies.," Figures \ref{fig:quality_lgm} and \ref{fig:quality_lgm_nuvr} demonstrate that the galaxies in theHI sample have larger average sizes than both $C_{M*}$ and $C_{M*,NUV-r}$ control sample galaxies."435" The mean value of Απο of the HI sample is larger by 1.2 and 0.3 kKpe than the mean values of Απ of the Cy, and Ciara, control sample galaxies respectively."," The mean value of $R_{50}(i)$ of the HI sample is larger by 1.2 and 0.3 kpc than the mean values of $R_{50}(i)$ of the $C_{M*}$ and $C_{M*,NUV-r}$ control sample galaxies respectively."436" In Figure 9.. we plot the relations between the half-light radius As, and MCHD/M. in four ditferent bins of stellar mass for galaxies in the HI sample."," In Figure \ref{fig:size4bands}, we plot the relations between the half-light radius $R_{50}$ and $/$ $_*$ in four different bins of stellar mass for galaxies in the HI sample."437 The mean values of MCHD/M. for each stellar mass bin are marked as crosses of ditferent sizes at the bottom of the plot., The mean values of $/$ $_*$ for each stellar mass bin are marked as crosses of different sizes at the bottom of the plot.438" These have been derived by Catinella et al (2010) using GASS survey galaxies for which HI masses have been measured down to a limiting MCHD/M, limit of ~1.5%.", These have been derived by Catinella et al (2010) using GASS survey galaxies for which HI masses have been measured down to a limiting $/$ $_*$ limit of $\sim 1.5 \%$.439 The majority of our HI-detected galaxies have atomic gas fractions that are above the average value — this is not surprising since galaxies detected in the ALFALFA survey make up the bulk of our sample., The majority of our HI-detected galaxies have atomic gas fractions that are above the average value – this is not surprising since galaxies detected in the ALFALFA survey make up the bulk of our sample.440" We see that the half-light radius Rsy measured in the e-band increases as a function of MIHD/M, at a given value of M.", We see that the half-light radius $_{50}$ measured in the $g$ -band increases as a function of $/$ $_*$ at a given value of $M_*$.441" This result is consistent with the sealing relations published in Zhang et al (2009) and CIO. which clearly showed that MCHD/M, correlates strongly with stellar surface density µ. (Le. with galaxy size at a fixed value of the stellar mass)."," This result is consistent with the scaling relations published in Zhang et al (2009) and C10, which clearly showed that $/$ $_*$ correlates strongly with stellar surface density $\mu_*$ (i.e. with galaxy size at a fixed value of the stellar mass)."442 Indeed. both studies find that Mt(HD/M. can be best predicted using a of colour and µ..," Indeed, both studies find that $/$ $_*$ can be best predicted using a of colour and $\mu_*$."443 As discussed in Zhang et al (2009). these scalings can be understood in terms of the Kennicutt-Schmidt law ofstar formation (Kennicutt[998).. which states that the surface density of star ormation scales with the surface density of cold gas as a with slope ~|.," As discussed in Zhang et al (2009), these scalings can be understood in terms of the Kennicutt-Schmidt law of star formation \citep{Kennicutt98}, which states that the surface density of star formation scales with the surface density of cold gas as a power-law with slope $\sim 1.4$."444 A star formation law of this form leads to the expectation that where a.b and c are constants.," A star formation law of this form leads to the expectation that where $a,b$ and $c$ are constants."445" The NUV-r colour is an excellent proxy for SFR/M, (especially in the blue sequence). so his leads to a prediction of a linear relation linking HI mass Traction. NUV-r colour and stellar surface mass density."," The $r$ colour is an excellent proxy for $M_*$ (especially in the blue sequence), so this leads to a prediction of a linear relation linking HI mass fraction, $r$ colour and stellar surface mass density."446" The two right hand panels of Figure 9. show that sizes of νε, and Cysovpy, Control galaxies exhibit the same increase as for the HI-detected galaxies."," The two right hand panels of Figure \ref{fig:size4bands} show that sizes of $C_{M*,NUV-r}$ and $C_{M_*,NUV-r,\mu_*}$ control galaxies exhibit the same increase as for the HI-detected galaxies."447 This supports our hypothesis that the scaling of galaxy size with HI mass fraction arises as a consequence of the star formation rate-gas surface density relation., This supports our hypothesis that the scaling of galaxy size with HI mass fraction arises as a of the star formation rate-gas surface density relation.448 In Figure 10.. we analyze trends in the ratio of the o-band and i-band half-light radii as a function of HI mass fraction.," In Figure \ref{fig:sizeratio4bands}, we analyze trends in the ratio of the $g$ -band and $i$ -band half-light radii as a function of HI mass fraction."449 R50 defined in the e-band is always slightly larger than R50 defined in the i-band. because the light from younger stellar populations is generally spread over a larger ettective radius than the light from older stellar populations.," R50 defined in the $g$ -band is always slightly larger than R50 defined in the $i$ -band, because the light from younger stellar populations is generally spread over a larger effective radius than the light from older stellar populations."450 The ratio between the ¢ and i-band radii increases as a function of gas fraction. and is also larger for more massive galaxies at a fixed value of MOHD/M.," The ratio between the $g$ and $i$ -band radii increases as a function of gas fraction, and is also larger for more massive galaxies at a fixed value of $M_*$ ."451. As we will show in the next section. this difference in e- and i-band half-light radii is also found when we analyze ο—i colour gradients.," As we will show in the next section, this difference in $g$ - and $i$ -band half-light radii is also found when we analyze $g-i$ colour gradients."452 In the panels to the right. we plot sizeditferences for the three control samples.," In the panels to the right, we plot sizedifferences for the three control samples."453" Neither the Cy, nor the νε, control samples show similar"," Neither the $C_{M*}$ nor the $C_{M*,NUV-r}$ control samples show similar"454primarily distributed rather uniformly at all stellar longitudes.,primarily distributed rather uniformly at all stellar longitudes.455" Spatially unevenly distributed X-ray emitting regions at low latitudes contribute at most We note, that some additional contribution from high latitudes or extended regions to the X-ray emission is not in contradiction with the observed light curves."," Spatially unevenly distributed X-ray emitting regions at low latitudes contribute at most We note, that some additional contribution from high latitudes or extended regions to the X-ray emission is not in contradiction with the observed light curves."456" Overall, the X-ray emitting corona of Altair needs to be rather stable on timescales of several weeks and as shown bythe previous ray detections, also on timescales of many years."," Overall, the X-ray emitting corona of Altair needs to be rather stable on timescales of several weeks and as shown bythe previous X-ray detections, also on timescales of many years."457 Intrinsic changes of Altair's surface features might also contribute to the variability of its X-ray brightness., Intrinsic changes of Altair's surface features might also contribute to the variability of its X-ray brightness.458" If this variability is due to weaker unresolved flaring activity as expected for stellar coronae, it should affect the spectral properties of the observed X-ray emission."," If this variability is due to weaker unresolved flaring activity as expected for stellar coronae, it should affect the spectral properties of the observed X-ray emission."459" Specifically, a spectral hardening during the X-ray brighter phases should be present; caused by the higher temperature plasma in activity related, X-ray emitting features like active regions or small, unresolved flares."," Specifically, a spectral hardening during the X-ray brighter phases should be present; caused by the higher temperature plasma in activity related, X-ray emitting features like active regions or small, unresolved flares."460" To investigate this possibility, we derive from the PN data for each time-bin a hardness ratio and compare it to the corresponding X-ray brightness."," To investigate this possibility, we derive from the PN data for each time-bin a hardness ratio and compare it to the corresponding X-ray brightness."461 We use the hardness ratio HR=(H—S)/(H+S) with S —00.6 keV and H —22.0 keV being the respective photon energy bands., We use the hardness ratio $HR=(H-S)/(H+S)$ with $S$ 0.6 keV and $H$ 2.0 keV being the respective photon energy bands.462" The spectral range in our soft band mainly covers the X-ray emission from cooler (< 2MK) plasma, the harder band traces the hotter plasma."," The spectral range in our soft band mainly covers the X-ray emission from cooler $\lesssim$ 2MK) plasma, the harder band traces the hotter plasma."463" In the lower panel of refpha,rweshowthethusobtainedspectralhardnessvs.X−Asa H", In the lower panel of \\ref{pha_hr} we show the thus obtained spectral hardness vs. X-ray brightness relation for Altair and overlaid linear regression curves for each observation; the errors (not shown) on the individual data point are around 0.01 (Rate) and 0.05 $HR$ ).464o," The linear regressions are both positive, supporting the picture that enhanced magnetic activity contributes to the observed variability."465weve," Combining the data from both observations, we derive slope of $0.54\pm0.39$."466"r, compared to active stars the slope is rather flat and the scatter is also quite large."," However, compared to active stars the slope is rather flat and the scatter is also quite large."467" While extreme spectral variations were not expected given the observed moderate variability, another cause of variability that does not produce significant spectral changes needs to be present."," While extreme spectral variations were not expected given the observed moderate variability, another cause of variability that does not produce significant spectral changes needs to be present."468" Altogether, our findings confirm the presence of a corona due to magnetic activity."," Altogether, our findings confirm the presence of a corona due to magnetic activity."469" It is generated by a dynamo that supposable operates in the thin outer convective layer of Altair, predominantly at equatorial up to intermediate latitudes."," It is generated by a dynamo that supposable operates in the thin outer convective layer of Altair, predominantly at equatorial up to intermediate latitudes."470" Altair’s corona is overall stable, whereas rotational modulation, transient coronal features as well as intrinsic variability due to magnetic activity contribute to the moderate changes of its ray brightness."," Altair's corona is overall stable, whereas rotational modulation, transient coronal features as well as intrinsic variability due to magnetic activity contribute to the moderate changes of its X-ray brightness."471" To study the coronal plasma properties of Altair and its variability, we performed a global spectral analysis, i.e., we modelled the full spectrum of each detector in the spectral range where sufficient source signal is present."," To study the coronal plasma properties of Altair and its variability, we performed a global spectral analysis, i.e., we modelled the full spectrum of each detector in the spectral range where sufficient source signal is present."472 We first investigated the spectra from the two observations separately and found only marginal differences., We first investigated the spectra from the two observations separately and found only marginal differences.473" As an example we show in refpnspec the two PN spectra, which are obviously very similar."," As an example we show in \\ref{pnspec} the two PN spectra, which are obviously very similar."474 Therefore we determined the global X-ray properties of Altair from all exposures combined and applied one spectral model to describe all RGS+MOS spectra (the RGS spectrum is shown in refrgs))., Therefore we determined the global X-ray properties of Altair from all exposures combined and applied one spectral model to describe all RGS+MOS spectra (the RGS spectrum is shown in \\ref{rgs}) ).475 We examined various multi-temperature models and found that a three temperature model is most suitable to describe the spectra., We examined various multi-temperature models and found that a three temperature model is most suitable to describe the spectra.476" We also fitted the high resolution data alone, especially to check the elemental abundances."," We also fitted the high resolution data alone, especially to check the elemental abundances."477" In this procedure Mg and Si that were taken from the combined RGS+MOS fit since no significant features below, that is where these elements produce strong X-ray lines, are present in the RGS spectrum."," In this procedure Mg and Si that were taken from the combined RGS+MOS fit since no significant features below, that is where these elements produce strong X-ray lines, are present in the RGS spectrum."478 A two-temperature model is sufficient to describe the RGS data alone and we find overall consistent results with the combined fit., A two-temperature model is sufficient to describe the RGS data alone and we find overall consistent results with the combined fit.479" The coronal properties derived from both fitting procedures, i.e. abundances relative to solar values, temperatures and emission measures as well as the corresponding X-ray luminosity, are given in Table 2.."," The coronal properties derived from both fitting procedures, i.e. abundances relative to solar values, temperatures and emission measures as well as the corresponding X-ray luminosity, are given in Table \ref{par}."480" To determine Altair’s X-ray activity level, that is described by the Lx/Lyo ratio, we additionally calculated the bolometric luminosity from its visual magnitude."," To determine Altair's X-ray activity level, that is described by the $L_{\rm X}/L_{\rm bol}$ ratio, we additionally calculated the bolometric luminosity from its visual magnitude."481" We find Ly.=3.8x1034 erg/s, thus Altair is roughly ten times brighter than the Sun."," We find $L_{\rm bol}= 3.8 \times 10^{34}$ erg/s, thus Altair is roughly ten times brighter than the Sun."482" This value indicates that Altair is slightly evolved, but closer to the main sequence than to the subgiant luminosity class, consistent with the isochrone based age estimation of about GGyr (?).."," This value indicates that Altair is slightly evolved, but closer to the main sequence than to the subgiant luminosity class, consistent with the isochrone based age estimation of about Gyr \citep{lac99}."483" The X-ray luminosity is given in the 0.2—22.0 keV band, where the EPIC instruments provided useful data."," The X-ray luminosity is given in the 2.0 keV band, where the EPIC instruments provided useful data."484" When adopting for comparison the ROSAT band 22.4 keV), the corresponding flux is roughly crosscheck we also derived X-ray fluxes for each detector other."," When adopting for comparison the ROSAT band 2.4 keV), the corresponding flux is roughly As a crosscheck we also derived X-ray fluxes for each detector independently and find that they overall agree among each other."485" Discrepancies on Lx are at the the PN detector predicting a slightly higher flux compared to RGS and MOS, especially at energies below kkeV. The very low activity level of logLx/Lpo.= —7.4, a value that is several ten thousand times below those of active low-mass"," Discrepancies on $L_{\rm X}$ are at the the PN detector predicting a slightly higher flux compared to RGS and MOS, especially at energies below keV. The very low activity level of $\log L_{\rm X}/L_{\rm bol}= -7.4$ , a value that is several ten thousand times below those of active low-mass"486stars in the images to achieve the most accurate alignment »ossible.,stars in the images to achieve the most accurate alignment possible.487 Unfortunately. the field of view of our images is small and in most cases we could only use the nucleus of he galaxy itself to determine the reference. position.," Unfortunately, the field of view of our images is small and in most cases we could only use the nucleus of the galaxy itself to determine the reference position."488 The individual images were thus registered and shifted. to a common position (in all cases accurate to a fraction of the pixel size) and averaged., The individual images were thus registered and shifted to a common position (in all cases accurate to a fraction of the pixel size) and averaged.489 In the process. the pixel size was valved to increase sampling (reduced [rom 07248 to 07124).," In the process, the pixel size was halved to increase sampling (reduced from $\farcs 248$ to $\farcs490124$ )."491 Finally. we rotated the images by an angle of SS.S in order o obtain the correct north (up). east (left) orientation.," Finally, we rotated the images by an angle of $\deg$ in order to obtain the correct north (up), east (left) orientation."492 Although we subtracted the background. sky as described in Section 2.3. we cannot check on how accurate our sky subtraction is since the skv is variable. and the [rame is small compared. with the galaxy.," Although we subtracted the background sky as described in Section 2.3, we cannot check on how accurate our sky subtraction is since the sky is variable, and the frame is small compared with the galaxy."493 TFhis makes it very hard. to estimate to what level the [lux registered. in the photometric aperture is inlluenced by any residual background emission., This makes it very hard to estimate to what level the flux registered in the photometric aperture is influenced by any residual background emission.494 For that reason. we have used aperture photometry from the literature.," For that reason, we have used aperture photometry from the literature."495 By calibrating with multiple aperture measurements we can better estimate the value of the sky and the ellicieney of the system., By calibrating with multiple aperture measurements we can better estimate the value of the sky and the efficiency of the system.496 ΙΓ only one useful aperture is available. we can use onlv the relatively right part of the image.," If only one useful aperture is available, we can use only the relatively bright part of the image."497 We were able to find the aperture yhotometry in JJ. 44 ancl dv-bancls for all but one of he galaxies. NGC 1530.," We were able to find the aperture photometry in $J$, $H$ and $K$ -bands for all but one of the galaxies, NGC 1530."498 For this galaxy. we used. the average magnitude olfset as determined for the rest. of he galaxies.," For this galaxy, we used the average magnitude offset as determined for the rest of the galaxies."499 Literature photometry sources include Class (1976). Aaronson (1977). AleAlary. AIeLaren Crabtree (1979). Dalzano Weecman (1981). Willner ct al. (," Literature photometry sources include Glass (1976), Aaronson (1977), McAlary, McLaren Crabtree (1979), Balzano Weedman (1981), Willner et al. ("5001985) and Spinoglio ct al. (,1985) and Spinoglio et al. (5011995).,1995).502 To check the quality of our photometry. we also made comparisons with data from the Literature.," To check the quality of our photometry, we also made comparisons with data from the literature."503 We. selected values from Glass (1984). Cidziel. Wyna-Williams Becklin," We selected values from Glass (1984), Cidziel, Wynn-Williams Becklin"504 a," \citep[see][for a review]{renzini06}. \citep{kodama97,kauffmann98,bernardi05}. \citep[e.g.,][]{faber07,mei09}."505s it allows color measurements even for faint ealaxics. and robust morphological measurements. which mclude either manual classification of E. S0. Sp. aud hr (e.g.Postmanetal.2005). orautomated separation of E/S0 types from later types (0.8...Abrahametal.2007).," as it allows color measurements even for faint galaxies, and robust morphological measurements, which include either manual classification of E, S0, Sp, and Irr \citep[e.g.,][]{postman05} orautomated separation of E/S0 types from later types \citep[e.g.,][]{abraham07}."506. Iu particular. combined color and morphological data cau be used to measure the color scatter of the E/SO galaxics in the CAIR. which places constraints ou the formation epochs of stars in cluster ealaxics (e.g...Boweretal. 1992).," In particular, combined color and morphological data can be used to measure the color scatter of the E/S0 galaxies in the CMR, which places constraints on the formation epochs of stars in cluster galaxies \citep[e.g.,][]{bower92b}."507 Until now. 21 systematic space-based studies of the CNIB in the cores of galaxy clusters havebeen couducted almost) exclusivelv using a handful of XN-raw selected ealaxy clusters (vauDokkietal.1998.2000:Blakesleeal. 2009).," Until now, $z \sim 1$ systematic space-based studies of the CMR in the cores of galaxy clusters havebeen conducted almost exclusively using a handful of X-ray selected galaxy clusters \citep{vandokkum98,vandokkum00,blakeslee03,mei06,blakeslee06,mei09}."508. The couseusus picture from these studies is +iat stars in these ealaxies were formed at +=2., The consensus picture from these studies is that stars in these galaxies were formed at $z \gtrsim 2$.509 Iu this Letter. we preseut the first precision statistical mcasurements of the scatter of the CATR in au optically-selected sample of galaxy clusters at 2~1d.," In this Letter, we present the first precision statistical measurements of the scatter of the CMR in an optically-selected sample of galaxy clusters at $z\sim1$."510" Where necessary. we adopt a flat. Q,,=0.30 cosinology with Il =70lans | |l."," Where necessary, we adopt a flat, $\Omega_m=0.30$ cosmology with $H_0=70$ km $^{-1}$ $^{-1}$."511 huaeiug for this study was acquired as part of the Clustey: SN. Survey Program (Program. Number 10196. PI: Perhuutter).," Imaging for this study was acquired as part of the Cluster SN Survey Program (Program Number 10496, PI: Perlmutter)."512 Nine optically-sclected clusters were dmaeed at various telescope roll augles for a total of 5000-16000s with the Advanced Camera for Surveys (ACS) on unultiple visits 3752-5008 per sub-exposure) in του (FShOLP) aud for LOQ0-8000s in fees (CFTT5W) (Dawsonetal.2009)., Nine optically-selected clusters were imaged at various telescope roll angles for a total of 5000-16000s with the Advanced Camera for Surveys (ACS) on multiple visits (375s-500s per sub-exposure) in $z_{850}$ (F850LP) and for 1000-8000s in $i_{775}$ (F775W) \citep{dawson09}.513. Eight clusters are derived frou. the Red-Sequeuce. Cluster Survey (RCS.Cdadders&Yee 2005).. aud the niuth. XLSS J0223.0|0136. was originally included in this sample as au IR-detected cluster from the SpARCS survey (Muzziuetal.2008: 2008).. but had previously been detected iu both the IR (Audreouetal.2005) aud N-ray (Pierre 200L)..," Eight clusters are derived from the Red-Sequence Cluster Survey \citep[RCS,][]{gladders05}, and the ninth, XLSS J0223.0+0436, was originally included in this sample as an IR-detected cluster from the SpARCS survey \citep{muzzin08,wilson08}, , but had previously been detected in both the IR \citep{andreon05} and X-ray \citep{pierre04}. ."514 We consider it here as an optical-IR, We consider it here as an optical-IR515In Fie.l the NVSS-anap of 2.7 at 1.1 CIIz is shown.,In \ref{map} the NVSS-map of $-$ 2.7 at 1.4 GHz is shown.516" Here the maps are plotted with the ""Skvview oikage (Ebert et al. 1998)).", Here the maps are plotted with the “Skyview” package (Ebert et al. \cite{sky}) ).517" The circular shell structure of the SNR has the aueular diameter Έτος(ήν, while the oeiteeral fiux density in the map. S,=070.3 Jv. is mach ower than the value extrapolated from the spectrum (sec low)."," The circular shell structure of the SNR has the angular diameter $17.0\pm0.2$, while the integral flux density in the map, $_\nu = 0.7\pm0.3$ Jy, is much lower than the value extrapolated from the spectrum (see below)."518 The large uncertainty of the NWSS flux depends strouely ou the backeround level definition aud could only ο a lower linut because he NVSS suffers frou lack of zero-spaciue data and its imiages are iuscusitive to suooth radio structures mich lavecr than several arcu., The large uncertainty of the NVSS flux depends strongly on the background level definition and could only be a lower limit because the NVSS suffers from lack of zero-spacing data and its images are insensitive to smooth radio structures much larger than several arcmin.519" The fux weighted centroid of the μ.ο... has the Galactic coordiuatco dDel671167 and hb= 276689 or equatorial ones: RADECLOSO = στης”, 009/550."," The flux weighted centroid of the source has the Galactic coordinates: $l$ 167 and $b=-2$ 689 or equatorial ones: RADEC1950 = $16^h27^m08^s$, $-$."520 Recently Gacusler (1998)) has investigated the nature of the the bilateral SNRs. one of them. G03.8-0.3. las a bilateral structure verv similar with 2.7.," Recently Gaensler \cite{bilat}) ) has investigated the nature of the the bilateral SNRs, one of them, G03.8-0.3, has a bilateral structure very similar with $-$ 2.7."521 Their surface brightuess. angular sizes are close.," Their surface brightness, angular sizes are close."522 We could estimate the value of c defined to be the acute augle between the sviunietry. axis of the SNR aud the Calactic plane., We could estimate the value of $\psi$ defined to be the acute angle between the symmetry axis of the SNR and the Galactic plane.523 We ft the svuunetrv axis using only the bright circular arcs., We fit the symmetry axis using only the bright circular arcs.524 This gives a value c=177 x., This gives a value $\psi=17\degr\pm3\degr$ .525 Thus the svununnetry axis is aligued close to the Galactic plane (see Fig.1))., Thus the symmetry axis is aligned close to the Galactic plane (see \ref{map}) ).526 It is uot clear whether the ceutral weak &luneut with a brightuess of nearly 1.5 mJy/bemn (~3 σ) located close to the svuuuctry axis is real aud associated with this SNR or not., It is not clear whether the central weak filament with a brightness of nearly 1.5 mJy/beam $\sim3-4\sigma$ ) located close to the symmetry axis is real and associated with this SNR or not.527 The new radio mapping are needed., The new radio mapping are needed.528 The sources in the field of this map from the NVSS source catalog are imarked dy white crosses and the munbers around them annotate the fractional polarization in per cent., The sources in the field of this map from the NVSS source catalog are marked by white crosses and the numbers around them annotate the fractional polarization in per cent.529 In Table a list of the NVSS SOTLECCS with detectable linear polarization is given., In Table \ref{src} a list of the NVSS sources with detectable linear polarization is given.530 This table is obtained by the program in the astroplivsica catalogs data base CATS (Verkhodanov el al. 1997))., This table is obtained by the program in the astrophysical catalogs data base CATS (Verkhodanov el al. \cite{CATS}) ).531 We have reduced it to six columns: umber. coordinates (21950). flux density. polarized intensity or power o: polarization in per cent: the last column iudicates whether the source is a part of the SNR or not.," We have reduced it to six columns: number, coordinates (B1950), flux density, polarized intensity or power of polarization in per cent; the last column indicates whether the source is a part of the SNR or not."532 It isremarkable tha details Guclided iu the NVSS source catalog) of brigl western and castern ares of theshell are highly. polarized, It is remarkable that details (included in the NVSS source catalog) of bright western and eastern arcs of theshell are highly polarized533Infrared-Faint Radio Sources (IFRS) were first classified by ?.. who identified them as sources detected at GGHz using the the Australia Telescope Compact Array (ATCA). but absent in deep infrared images from the Wide-Area Infrared Extragalactic (SWIRE) survey (?)).,"Infrared-Faint Radio Sources (IFRS) were first classified by , who identified them as sources detected at GHz using the the Australia Telescope Compact Array (ATCA), but absent in deep infrared images from the Wide-Area Infrared Extragalactic (SWIRE) survey )."534 In the process of matching IR counterparts to radio sources from the Australia Telescope Large Area Survey (ATLAS). found a total of 22 radio sources to which no infrared counterpart could be reasonably identified in any of the observedSpitzer bands 4m— um). down to a sensitivity of στον= LOgJy.," In the process of matching IR counterparts to radio sources from the Australia Telescope Large Area Survey (ATLAS), found a total of 22 radio sources to which no infrared counterpart could be reasonably identified in any of the observed bands $\mu$ m– $\mu$ m), down to a sensitivity of $\sigma_{3.6\,\mu m}\,=\,1.0\,\mu$ Jy."535 A similar result was obtained by who found 31 IFRS in the ELAIS-SI at similar IR and radio sensitivities., A similar result was obtained by who found 31 IFRS in the ELAIS-S1 at similar IR and radio sensitivities.536 A substantial fraction of these IFRS have flux densities of several mJy at GGHz. and some are even found at mmJy.," A substantial fraction of these IFRS have flux densities of several mJy at GHz, and some are even found at mJy."537 A first attempt to investigate the nature of the IFRS discovered in the ATLAS survey was undertaken by?.. who observed two IFRS with Very Long Baseline Interferometry (VLBI). and detected one.," A first attempt to investigate the nature of the IFRS discovered in the ATLAS survey was undertaken by, who observed two IFRS with Very Long Baseline Interferometry (VLBI), and detected one."538 They argued that [IFRS are AGN-driven objects because the VLBI detection indicated a brightness temperatures in excess of 106 KK. Such high temperatures cannot be reached by thermal emission mechanisms and hence indicate non-thermal radio emission produced by an AGN., They argued that IFRS are AGN-driven objects because the VLBI detection indicated a brightness temperatures in excess of $10^6$ K. Such high temperatures cannot be reached by thermal emission mechanisms and hence indicate non-thermal radio emission produced by an AGN.539 Later?.. also using VLBI observations. confirmed that at least a fraction of the [IFRS population must contain an AGN contributing to the total IFRS spectral energy distribution.," Later, also using VLBI observations, confirmed that at least a fraction of the IFRS population must contain an AGN contributing to the total IFRS spectral energy distribution."540 These results were complemented by who identified an IPRS population in theSpirzer extragalactic First Look Survey (XFLS)., These results were complemented by who identified an IFRS population in the extragalactic First Look Survey (xFLS).541 Modelling their spectral energy distributions (SEDs) with template SEDs from known objects to estimate IFRS redshifts. they found that the characteristics of IFRS resembled a variety of 3C sources redshifted to € zzx55.," Modelling their spectral energy distributions (SEDs) with template SEDs from known objects to estimate IFRS redshifts, they found that the characteristics of IFRS resembled a variety of 3C sources redshifted to $\leq$ $\leq$ 5."542 A more detailed SED analysis was presented by?.. who used new. very deep data in the Deep Field South (CDF-S) field obtained during two legacy programs. SIMPLE (Damen et al..," A more detailed SED analysis was presented by, who used new, very deep data in the Deep Field South (CDF-S) field obtained during two legacy programs, SIMPLE (Damen et al.,"543 in prep.), in prep.)544 and FIDEL (PI: Dickinson)., and FIDEL (PI: Dickinson).545" With a sensitivity of ce,=— O.l64dy. IR counterparts for two of the four investigated [IFRS were identified."," With a sensitivity of $\sigma_{3.6\,\rm{\mu546 m}}\,=\,0.16\,\mu$ Jy, IR counterparts for two of the four investigated IFRS were identified."547 Furthermore. a faint optical counterpart to one of the IFRS was serendipitously identified within the GOODS HST/ACS imaging campaign(2).," Furthermore, a faint optical counterpart to one of the IFRS was serendipitously identified within the GOODS HST/ACS imaging campaign."548". A. second IFRS located within the GOODS-South field was found to have no optical detection at all (implying (, AAB—mag). puttingstrongconstraintsontheirS EDs."," A second IFRS located within the GOODS-South field was found to have no optical detection at all (implying $>$ AB-mag), putting strong constraints on their SEDs."549Huynhetal., Huynh et al.550alsosearchedoti wavelengthsdatasetsintheC DF− forcounterparts. inparticulartheChandra2," also searched other deep multi-wavelengths data sets in the CDF-S for counterparts, in particular the Ms source catalog, the GEMS HST imaging campaign and the MUSYC catalog."551 Mssourcecatalog(?). . th« ravortheoptical/N IRregime., These searches did not yield any counterparts in the X-ray or the optical/NIR regime.552ModellingtemplateS EDstorepresenttherad likeob jecteanre producethedatawhenred shiftedtoz JXIR--detectionyandz» 44. (IR. non-detection).," Modelling template SEDs to represent the radio and, if available, the IR and optical detections, found that a 273-like object can reproduce the data when redshifted to 2 (IR-detection) and $>$ 4 (IR non-detection)."553 The IR. non-detections could not be explained by any template SED at redshifts smaller than ~4., The IR non-detections could not be explained by any template SED at redshifts smaller than $\sim$ 4.554 These estimates constrain the redshift range of IFRS. placing at least a significant. fraction of them at Z¢44.," These estimates constrain the redshift range of IFRS, placing at least a significant fraction of them at $>$ 4."555Ontheotherhand. giventhatnoneofthel FRS wasdetectedat24 um. conclude that IFRS fall well beyond the IR-radio correlation /Scoan) parameter of 0.84 would imply a 24m flux density of 7mmJy for a typical IFRS with Sio=| mJy|Appleton2004.," On the other hand, given that none of the IFRS was detected at $\mu$ m, conclude that IFRS fall well beyond the IR-radio correlation ) parameter of 0.84 would imply a $\mu$ m flux density of mJy for a typical IFRS with $_{20\,\rm{cm}}\,=\,1\,$ ."556 Hence the IFRS radio emission. can not be explained by star-forming processes but is likely to be produced by AGN., Hence the IFRS radio emission can not be explained by star-forming processes but is likely to be produced by AGN.557 This finding is supported by applying various calibrations of the GGHz luminosity as star formation rate (SFR) tracer., This finding is supported by applying various calibrations of the GHz luminosity as star formation rate (SFR) tracer.558 For instance. the classical calibration by gives star formation rates of around a million solar nasses per year for a typical IFRS assumed to. be located at 44. which is unphysical and hence suggests that the radio emission of IFRS cannot be only caused by star formation.," For instance, the classical calibration by gives star formation rates of around a million solar masses per year for a typical IFRS assumed to be located at 4, which is unphysical and hence suggests that the radio emission of IFRS cannot be only caused by star formation."559 Using new radio data spanning GGHz to GGHz. analysed the radio properties of 18 ATLAS IFRS.," Using new radio data spanning GHz to GHz, analysed the radio properties of 18 ATLAS IFRS."560 They found unusually steep radio spectral indices with a median α of —-L.4 (S.x v) and no spectral index flatter than —0.5.," They found unusually steep radio spectral indices with a median $\alpha$ of $-1.4$ $S\propto\nu^\alpha$ ), and no spectral index flatter than $-0.5$."561 According to the z-a relation that predicts steeper radio spectralindices at higher redshifts. these findings suggest that IFRS reside in the high-redshift Universe.," According to the $\alpha$ relation that predicts steeper radio spectralindices at higher redshifts, these findings suggest that IFRS reside in the high-redshift Universe."562 Another result from Middelberg et al., Another result from Middelberg et al.563 is that the ratio between radio and IR flux densities. S20 ¢n/S3.6ym. 18 much larger for IFRS than for the general radio source population. and potentially exceeds 107. a value which is similar to that found in a sample of high," is that the ratio between radio and IR flux densities, $S_{20\,\rm{cm}}/S_{3.6\,\rm{\mu m}}$ , is much larger for IFRS than for the general radio source population, and potentially exceeds $10^4$ , a value which is similar to that found in a sample of high"564Any point in the simulation volume can be specified as the beginning of a rav.,Any point in the simulation volume can be specified as the beginning of a ray.565 Currently. is configured to treat point sources whose properties are specified by an input file. recombination ravs from ionized SPIEL particles. and background. [fluxes by specifving points on the simulation volume walls as sources.," Currently, is configured to treat point sources whose properties are specified by an input file, recombination rays from ionized SPH particles, and background fluxes by specifying points on the simulation volume walls as sources."566 llere we review the recombination processes that. produce ionizing photons in a 11ο gas (e.g.?)..," Here we review the recombination processes that produce ionizing photons in a H/He gas \citep[e.g.,][]{1989agna.book.....O}."567 Ehe following bound transitions produce continuous spectra., The following free-bound transitions produce continuous spectra.568 These spectra can be caleulated exactly using the Milne relations., These spectra can be calculated exactly using the Milne relations.569 For Lvdrogen. the emission cocllicient for the above process is (?).. where cd is the photo absorption cocllicient of llvdrogen in the ground state.," For Hydrogen, the emission coefficient for the above process is \citep{1989agna.book.....O}, where $\sigma_{\rm H}^1$ is the photo absorption coefficient of Hydrogen in the ground state."570 Similar relations can be derived. for the other. [ree-bound. processes. however. in practice these highly peaked spectra can be approximated by delta funetions just above the appropriate threshold (in parentheses in the equations above)," Similar relations can be derived for the other free-bound processes, however in practice these highly peaked spectra can be approximated by delta functions just above the appropriate threshold (in parentheses in the equations above)."571 Following free-bound captures to excited Helium states. the following bound-bound. transitions can also produce ionizing photons.," Following free-bound captures to excited Helium states, the following bound-bound transitions can also produce ionizing photons."572 The various bound-bound transitions above have relative probabilities that depend on the environment (νου electron. density. temperature. ionization state) ancl so the weights to give to these processes should. be tailored to specific applications.," The various bound-bound transitions above have relative probabilities that depend on the environment (free electron density, temperature, ionization state) and so the weights to give to these processes should be tailored to specific applications."573 The most straight forward way to deal with this dilfuse radiation is to use the on-the-spot (OLS) approximation., The most straight forward way to deal with this diffuse radiation is to use the on-the-spot (OTS) approximation.574 The OTS approximation makes the assumption that recombination photons are absorbed in the vicinity. (the same SPILL particle) of the point where they are emitted., The OTS approximation makes the assumption that recombination photons are absorbed in the vicinity (the same SPH particle) of the point where they are emitted.575 Computationallv. this means no ray tracing is necessary for these photons.," Computationally, this means no ray tracing is necessary for these photons."576 For pure Hydrogen simulations. the OTS approximation amounts to using the reduced recombination rates in the appendix. (ease D).," For pure Hydrogen simulations, the OTS approximation amounts to using the reduced recombination rates in the appendix (case B)."577.. Here. one is making the assumption that cach electron capture directIy to the ground state produces a photon that tonizes a nearby Hydrogen atom and the two actions cllectively cancel one another.," Here, one is making the assumption that each electron capture directly to the ground state produces a photon that ionizes a nearby Hydrogen atom and the two actions effectively cancel one another."578 For simulations involving LHlelium. using the case D rates would be making the assumption that each Hell recombination to the ground state ionizes a nearby Hel atom while each Helle recombination to the ground state ionizes a nearby. Hell atom.," For simulations involving Helium, using the case B rates would be making the assumption that each HeII recombination to the ground state ionizes a nearby HeI atom while each HeIII recombination to the ground state ionizes a nearby HeII atom."579 This is a simple first approximation. but some of the Hell ground state captures will ionize HI. while some of the HoLLE ground state captures will ionize LIL and. Hel. To account for this would require a more detailed adjustment of the recombination ancl photoionization rates for the species involved.," This is a simple first approximation, but some of the HeII ground state captures will ionize HI, while some of the HeIII ground state captures will ionize HI and HeI. To account for this would require a more detailed adjustment of the recombination and photoionization rates for the species involved."580 The most computationally intensive option is to trace ravs lor each of these recombination processes and thereby take account of the fact that some of the photons will not be absorbed in the SPIE particle where they were ereated., The most computationally intensive option is to trace rays for each of these recombination processes and thereby take account of the fact that some of the photons will not be absorbed in the SPH particle where they were created.581 Again. the level of detail used. should be guided by the application at hand.," Again, the level of detail used should be guided by the application at hand."582 With it is possible to choose either the recombination ray or the OTS approach., With it is possible to choose either the recombination ray or the OTS approach.583 We plan to explore the accuracy of the Ο5 approximation in various geometries ancl densities in future work., We plan to explore the accuracy of the OTS approximation in various geometries and densities in future work.584 Ray tracing solutions to the radiative transfer problem solve the equation along 1D characteristics., Ray tracing solutions to the radiative transfer problem solve the equation along 1D characteristics.585 As such. an estimate of the optical depth along these characteristics is central to the problem.," As such, an estimate of the optical depth along these characteristics is central to the problem."586 In the SPILL formalism. a continuous density field. is represented. by a number of discrete [uid elements (particles) with smoothing lengths. h;.," In the SPH formalism, a continuous density field is represented by a number of discrete fluid elements (particles) with smoothing lengths $h_i$."587 These smoothing lengths are usually defined to keep a constant mass AM. inside the smoothing volume V;—aah}., These smoothing lengths are usually defined to keep a constant mass $M_{\rm sph}$ inside the smoothing volume $ V_i = \frac{4}{3} \pi h_i^3$.588 The properties of the fluid at any point are then estimated by averaging over all No particles in. the simulation weighted. by a smoothing kernel., The properties of the fluid at any point are then estimated by averaging over all $N$ particles in the simulation weighted by a smoothing kernel.589 In practice. one only averages over nearby particles. but this definition is equally valid and useful in the derivation to follow.," In practice, one only averages over nearby particles, but this definition is equally valid and useful in the derivation to follow."590 As an example. the density p(r;) at the position r; of the i particle is estimated as. where m; is the mass of the )u ‘nearest particleF and ο is a smoothing kernel.," As an example, the density $\rho(\mathbf{r}_i)$ at the position $\mathbf{r}_i$ of the $i^{\rm th}$ particle is estimated as, where $m_j$ is the mass of the $j^{\rm th}$ nearest particle and $W(r_{ij},h)$ is a smoothing kernel."591 An estimate of the Duid. property need. not be made at the position of a particle., An estimate of the fluid property need not be made at the position of a particle.592 An averaged value for any Iluid property can be defined. for an arbitrary point in space using two techniques., An averaged value for any fluid property can be defined for an arbitrary point in space using two techniques.593 One is the “scatter” method in which the desired. quantity is calculated. by averaging over every particle whose smoothing volume includes the point in question., One is the “scatter” method in which the desired quantity is calculated by averaging over every particle whose smoothing volume includes the point in question.594The study of the CeV-TeV conrponeut of esannununmna-ray bursts ds of ercat Huportance to understand the acceleration mechanisius and the sources plysical conditions.,The study of the GeV-TeV component of gamma-ray bursts is of great importance to understand the acceleration mechanisms and the sources physical conditions.595 The detection of GeV gamma-rays by ECRET diving some intense GRBs (Catellietab.1997). suggests the possibility. that a high cnerey component could be present iu all events.," The detection of GeV gamma-rays by EGRET during some intense GRBs \cite{egret}596 suggests the possibility that a high energy component could be present in all events."597 Furthermore several models predict CeWV and TeV emission. sometimes correlated with CHECRs production (see Baring1997.. for a review).," Furthermore several models predict GeV and TeV emission, sometimes correlated with UHECRs production (see \cite{baring}, for a review)."598 Due to the low fluxes aud the small sensitive areas of satellite experiments. eanunatravs of cucrey larecr than a few tens of GeV. mst be detected. by eround base experiments located at mountain altitude measuring the secondarv particles generated by ολαανν du the atimosphere.," Due to the low fluxes and the small sensitive areas of satellite experiments, gamma-rays of energy larger than a few tens of GeV, must be detected by ground based experiments located at mountain altitude measuring the secondary particles generated by gamma-rays in the atmosphere."599 At energies E « 10 TeV the number of particles reaching the erouik is to sinall to reconstruc the shower parameters usingo a standare air shower array. made of several detectors spread over large areas;," At energies E $<$ 10 TeV the number of particles reaching the ground is to small to reconstruct the shower parameters using a standard air shower array, made of several detectors spread over large areas."600 Ou tle contrary. a detector consisting of a full coverage laver of counters. providing a lieh eranunlarity sampling of all particle showers. can succestullv 1icasiure arrival direction and primary energv of sunuall showers. allowing the study of the unexplored range of eamuna chereies between 20 GeV and 300 GeV CAbbresciaetal. 1996)..," On the contrary, a detector consisting of a full coverage layer of counters, providing a high granularity sampling of all particle showers, can succesfully measure arrival direction and primary energy of small showers, allowing the study of the unexplored range of gamma energies between 20 GeV and 300 GeV \cite{proposal}. ."601mater]al located at the iuner edge of the torus.,material located at the inner edge of the torus.602 In either case. the scattering material is exposed to the intense nuclear [lux aud may produce siguilicant line emission aud. iu circumstances where the cli‘ect continuum Is COLipletely. blocked leaving only a baseline of the much weaker scattered COLiintuo. such lines cau iive very high equivalent widths (Chisellini19t 1)..," In either case, the scattering material is exposed to the intense nuclear flux and may produce significant line emission and, in circumstances where the direct continuum is completely blocked leaving only a baseline of the much weaker scattered continuum, such lines can have very high equivalent widths \citep{ghi94,kro94}."603 Tje spectral iudices fouid in Seyfert galaxies are distributed arouud a single value of Dzz1.9 (Naxda&Pounds1991:SuuithDone1996).," The spectral indices found in Seyfert galaxies are distributed around a single value of $\Gamma \approx 1.9$ \citep{np94,sd96}."604. Tie. best explanation for this is that mildly relalvistic thermal electrois or highly relativistic ou-thermal electrous Compton scatter UV phoOLs into the N-ray energy range (e.g. Sveussou 1996 aud references therein)., The best explanation for this is that mildly relativistic thermal electrons or highly relativistic non-thermal electrons Compton scatter UV photons into the X-ray energy range (e.g. Svensson 1996 and references therein).605 The rollover seen iu X-ray data above 100 keV lias focussed most atteujon ou therma inodels (Zdziarski 1996).," The rollover seen in X-ray data above 100 keV has focussed most attention on thermal models \citep{zdz95,gon96}."606.. A »opular scenario is tlie two-plase ciisk-coroia inodel. in which a hot X-ray corona Is located above a cold UV emitting accrelon cisk (Haardt&]araschi1991.1993).," A popular scenario is the two-phase disk-corona model, in which a hot X-ray corona is located above a cold UV emitting accretion disk \citep{hm91,hm93}."607. The UV. see photons for Compt«ii cooling of the euergeic electrous are produced from reprocessiug of the hard. X-ray spectrum iu he accretion disk., The UV seed photons for Compton cooling of the energetic electrons are produced from reprocessing of the hard X-ray spectrum in the accretion disk.608 Witl complete feedback. the approximate equipartition between the soft disk aicL hard. X-ray. Iluminuosiies leads naturally to Pz;1.9 for a wide rauge of optical deptlis (see also Stern et al.," With complete feedback, the approximate equipartition between the soft disk and hard X-ray luminosities leads naturally to $\Gamma \simgreat 1.9$ for a wide range of optical depths (see also Stern et al."609 1995)., 1995).610 However. or sources with spectra flatter thar Dal1.9. the clisk-corona model o. Haardt.&Maraschi(1991.1993) tmust be modified so that sone of the sol seed photous escape without being intercepted by the corona. tlis making for harder spectra als ewer photons are available for Comptou scatering (Haardtetal.1991).," However, for sources with spectra flatter than $\Gamma \approx 1.9$, the disk-corona model of \citet{hm91,hm93} must be modified so that some of the soft seed photons escape without being intercepted by the corona, thus making for harder spectra as fewer photons are available for Compton scattering \citep{haa94}."611. X-ray varlability studies of Sevfert 2 nuclei can provide importait clues to the processes by whcho N-rays are produced. as well as to the geometry of the circumuucear matter., X-ray variability studies of Seyfert 2 nuclei can provide important clues to the processes by which X-rays are produced as well as to the geometry of the circumnuclear matter.612 However. Varlability has been reported in ouly a few objects.," However, variability has been reported in only a few objects."613 Que example is MNrk 3. where the hard X-ray enmssion decreased. by a factor of two during a period of 3.6 years al. LODL)..," One example is Mrk 3, where the hard X-ray emission decreased by a factor of two during a period of 3.6 years \citep{awa90,mar92,iwa94}."614 Recent observations with have shown variability to exist ou tiue-scales of weeks (Gieorgantopotlosetal.1999)., Recent observations with have shown variability to exist on time-scales of weeks \citep{geo99}.615. The shape of the spectrum suggests that of tlie variability cau be attributed t«» elhiauges in the direct mur‘lear flux (Capοἱetal.1999)., The shape of the spectrum suggests that of the variability can be attributed to changes in the direct nuclear flux \citep{cap99}.616. Short teru variability (on time-scales of Lous) has been detected ist only oue object. NGC 19125 Uwasawaetal.1993:Guainazzi2000).," Short term variability (on time-scales of hours) has been detected in only one object, NGC 4945 \citep{iwa93,gua00}."617. These observations slippgest that à pa‘sec scale. geometrically hick molecular torus canuot be respousible for the bulk ‘the X-ray. absoption. (Mlaclejskietal.2Q00).," These observations suggest that a parsec scale, geometrically thick molecular torus cannot be responsible for the bulk of the X-ray absorption \citep{mad00}."618. Spectral variabiity has also )een claimed in a couple of Seylert 2 galaxies., Spectral variability has also been claimed in a couple of Seyfert 2 galaxies.619 For exam.jxle. the tsorption colum li NGC 7582 increased w MyoEx1077 ? between observations taken two vears apart (Nieetal.1998).," For example, the absorption column in NGC 7582 increased by $N_{\rm H} \sim 4 \times 10^{22}$ $^{-2}$ between observations taken two years apart \citep{xue98}."620. A similar variation in the absorber was suggested by Warwicsetal.(1993.. based oi a comparison betweenEinstein...EXOSAT.. aud observations.," A similar variation in the absorber was suggested by \citet{war93}, based on a comparison between, and observations."621" More 'ecent observations witl of this particular galaxy have shown tliat the absorber nay have a complex spatial st""ucture with a large column density (Ny,~107! em 7) covering of ha«d X-ray continuuur ald a smaller. but variable column completely covering the source (Turneretal. 2000)."," More recent observations with of this particular galaxy have shown that the absorber may have a complex spatial structure with a large column density $N_{\rm H} \sim 10^{24}$ $^{-2}$ ) covering of hard X-ray continuum, and a smaller, but variable column completely covering the source \citep{tur00}."622.As a further example. in NCC 7172. the spectral index. of the," As a further example, in NGC 7172, the spectral index of the"623Active Galactic Nuclei are powered by accretion onto a central massive black hole.,Active Galactic Nuclei are powered by accretion onto a central massive black hole.624 The gas surrounding the nucleus. some of which provides the fuel for the accretion process. often obscures it from direct view.," The gas surrounding the nucleus, some of which provides the fuel for the accretion process, often obscures it from direct view."625 Indeed. the hard shape of the spectrum of the X-ray Background argues that most accretion onto galactic nuclei is obscured citep| 9Y9MNRAS.303L..34F..," Indeed, the hard shape of the spectrum of the X-ray Background argues that most accretion onto galactic nuclei is obscured \\citep{1999MNRAS.303L..34F}."626 This is confirmed by deep Chandra and XMM-Newton imaging of the Sky with many AGN found to lie behind a column density of 1077.—107*em7 or more ApJS..1 citealt200239..3696:: citealt2005 ARAA..43..827B)})., This is confirmed by deep Chandra and XMM-Newton imaging of the Sky with many AGN found to lie behind a column density of $10^{22} - 10^{23} \pcmsq$ or more \\citealt{2002ApJS..139..369G}; \\citealt{2005ARA&A..43..827B}) ).627 Much of the obscuring material must Hie within the inner 100 pe. or its total mass would be prohibitive (see citealt2007astro.ph..1109M for a review).," Much of the obscuring material must lie within the inner 100 pc, or its total mass would be prohibitive (see \\citealt{2007astro.ph..1109M} for a review)."628 It is therefore part of the inner bulge of the host |galaxy., It is therefore part of the inner bulge of the host galaxy.629 Stars can also form from this gus. giving a nuclear starburst.," Stars can also form from this gas, giving a nuclear starburst."630 The gas is subject to the radiation pressure of the AGN. and can be ejected from the bulge if the nucleus becomes too bright.," The gas is subject to the radiation pressure of the AGN, and can be ejected from the bulge if the nucleus becomes too bright."631 Such AGN feedback may thereby remove the gas which fuels the nucleus and from which new stars form. so terminating the growth of both the central black hole and its host bulge.," Such AGN feedback may thereby remove the gas which fuels the nucleus and from which new stars form, so terminating the growth of both the central black hole and its host bulge."632 Simple caleulations of when this occurs citealtl1999MNRAS.308L..39F:: citealt2002MNRAS.329L..| SF: citealt2003ApJ...596L..?7K.. ApJ...618..569M:: citealt2005 citealt2006MNRAS.373L..16F))lead to the following relation between the mass of the black hole A/5g and the velocity dispersion of the bulge e: Assuming a gas fraction fo~0.1. this gives an Mpg—0 relation in good agreement with observations citealtl995ARAA..33..58I citealtl998AJ....| [5.2285M: K::citealt2000ApJ...539L...1 3G:: citealt200| ApJ...555L..79Fp.," Simple calculations of when this occurs \\citealt{1999MNRAS.308L..39F}; \\citealt{2002MNRAS.329L..18F}; \\citealt{2003ApJ...596L..27K}, , \\citealt{2005ApJ...618..569M}; \\citealt{2006MNRAS.373L..16F}) )lead to the following relation between the mass of the black hole $M_{\rm BH}$ and the velocity dispersion of the bulge $\sigma$: Assuming a gas fraction $f\sim 0.1$, this gives an $M_{\rm BH} -633\sigma$ relation in good agreement with observations \\citealt{1995ARA&A..33..581K}; \\citealt{1998AJ....115.2285M}; \\citealt{2000ApJ...539L..13G}; \\citealt{2001ApJ...555L..79F}) )."634 The limit when radiation pressure ejects mass is an effective Eddington limit relying on absorption of radiation by dust. not electron scattering.," The limit when radiation pressure ejects mass is an effective Eddington limit relying on absorption of radiation by dust, not electron scattering."635" The radiation pressure is amplified or boosted by a factor ;À which is the ratio of the effective. frequency weighted. absorption cross section for dusty gas. a, to that for electrons alone: op."," The radiation pressure is amplified or boosted by a factor $A$ which is the ratio of the effective, frequency weighted, absorption cross section for dusty gas, $\sigma_{\rm d}$ to that for electrons alone: $\sigma_{\rm T}$."636 The dust absorption is greatest in the UV and the value of <4 depends on the spectrum of the ΔΟΝ., The dust absorption is greatest in the UV and the value of $A$ depends on the spectrum of the AGN.637 The X-ray emission from the nucleus ensures that the gas and dust remain weakly ionized. and are effectively coupled by Coulomb forces so that pressure on the grains is shared with the surrounding gas.," The X-ray emission from the nucleus ensures that the gas and dust remain weakly ionized, and are effectively coupled by Coulomb forces so that pressure on the grains is shared with the surrounding gas."638 Boost factors computed for a standard AGN spectrum. using the radiation code are shown in and Fig.," Boost factors computed for a standard AGN spectrum, using the radiation code are shown in \\cite{2006MNRAS.373L..16F} and Fig."639 |., 1.640 They range from several hundred for low column densities and drop as the column density of gas increases until they approach unity when the gas becomes, They range from several hundred for low column densities and drop as the column density of gas increases until they approach unity when the gas becomes641anisotropic conduction.,anisotropic conduction.642 The solid black curve shows results from the weak turbulence driving case., The solid black curve shows results from the weak turbulence driving case.643 The total energy injection due to stirring motions is much smaller then the thermal energy content of the ICM., The total energy injection due to stirring motions is much smaller then the thermal energy content of the ICM.644" The gas velocity dispersion in the final state is ~50 km/s, ie. much smaller the the sound speed in the ICM."," The gas velocity dispersion in the final state is $\sim 50$ km/s, i.e., much smaller the the sound speed in the ICM."645" Such weak stirring motions are entirely conceivable and are consistent with, for example, repeated minor mergers, major mergers that stir the gas on large scales and cascade to smaller scales, AGN-driven motions or stirring by galaxy motions."," Such weak stirring motions are entirely conceivable and are consistent with, for example, repeated minor mergers, major mergers that stir the gas on large scales and cascade to smaller scales, AGN-driven motions or stirring by galaxy motions."646" For instance, ? find from cosmological simulations that even relatively relaxed clusters have internal velocities ~20% of the internal sound speed; ? finds turbulent velocities ~100—200kms~! in numerical simulations of gravitational wakes of galaxies in clusters."," For instance, \citet{nagai03} find from cosmological simulations that even relatively relaxed clusters have internal velocities $\sim 20\%$ of the internal sound speed; \citet{kim07} finds turbulent velocities $\sim 100-200 \, {\rm km \, s^{-1}}$ in numerical simulations of gravitational wakes of galaxies in clusters."647" The left panel demonstrates that the gas velocity field becomes preferentially tangential in this case, while the middle panel shows that the magnetic field is close to isotropic."," The left panel demonstrates that the gas velocity field becomes preferentially tangential in this case, while the middle panel shows that the magnetic field is close to isotropic."648 'This magnetic field topology is consistent with the right panel in Figure 2., This magnetic field topology is consistent with the right panel in Figure 2.649 The isotropization of the magnetic field leads to a significant boost in the level of the effective thermal conduction as shown in the right panel of Figure 3., The isotropization of the magnetic field leads to a significant boost in the level of the effective thermal conduction as shown in the right panel of Figure 3.650" In order to better understand the observed trends in the velocity, magnetic field and the effective conduction, we performed MHD runs with stirring but without conduction."," In order to better understand the observed trends in the velocity, magnetic field and the effective conduction, we performed MHD runs with stirring but without conduction."651 An example of the result from one of these runs is shown as a solid purple line in all three panels in Figure 3., An example of the result from one of these runs is shown as a solid purple line in all three panels in Figure 3.652 It is evident that the stirring leads to preferentially tangential velocity field., It is evident that the stirring leads to preferentially tangential velocity field.653 This behavior is very similar to the case when both the stirring and HBI operate., This behavior is very similar to the case when both the stirring and HBI operate.654 The reason for the preferentially tangential gas motions in the absence of conduction is that the characteristic driving frequency is lower then the local Brunt-Vaiisalla frequency wey., The reason for the preferentially tangential gas motions in the absence of conduction is that the characteristic driving frequency is lower then the local Brunt-Väiisällä frequency $\omega_{BV}$.655" In this case, gas motions excite gravity waves that become trapped within the radius where (g-modes)“turbulence frequency""σ/λSwey, where o and A are the gas velocity dispersion and the characteristic eddy size, respectively."," In this case, gas motions excite gravity waves $g$ -modes) that become trapped within the radius where “turbulence $\sim\sigma/\lambda\la\omega_{BV}$, where $\sigma$ and $\lambda$ are the gas velocity dispersion and the characteristic eddy size, respectively."656" Consequently, the distribution of the orientation of magnetic field versors is also tangential (see middle panel)."," Consequently, the distribution of the orientation of magnetic field versors is also tangential (see middle panel)."657 One subtle point worth emphasizing is that the magnetic fields have “memory” of the the fluid displacement history., One subtle point worth emphasizing is that the magnetic fields have “memory” of the the fluid displacement history.658" In other words, for weak fields, magnetic fields behave as “tracer particles” and, thus,"," In other words, for weak fields, magnetic fields behave as “tracer particles” and, thus,"659J=4 3cmission peaks at 53.1 witha PWLAL Av ~5A ,J=4–3 emission peaks at $\sim53.1$ with a FWHM $\Delta$ v $\sim5.4$ .660‘Takinginto account that the — emission is optical thinner than the emission ancl peaks at a velocity close to that of the — dip. we conclude that this emission indeed appears self-absorbed.," Takinginto account that the $^{13}$ $^{+}$ emission is optical thinner than the $^{+}$ emission and peaks at a velocity close to that of the $^{+}$ dip, we conclude that this emission indeed appears self-absorbed."661 This kind of spectral feature might be revealing the existence of a density eracient in the clump (Lliramatsuetal.2007).. consistent with the presence of an embedded: central source. with outfTowing activity.," This kind of spectral feature might be revealing the existence of a density gradient in the clump \citep{hira07}, consistent with the presence of an embedded central source with outflowing activity."662 LO is known that such molecular species enhances in molecular outllows (tawlinesetal.2004)., It is known that such molecular species enhances in molecular outflows \citep{raw04}.663. Ln ellect. a strong enhancement of the abundance is expected to occur in the boundary [aver between the outflow jet and the surrounding molecular core.," In effect, a strong enhancement of the $^{+}$ abundance is expected to occur in the boundary layer between the outflow jet and the surrounding molecular core."664 This would be due to the liberation and photoprocessing by the shock of the molecular material stored in the icy mantles of the dust., This would be due to the liberation and photoprocessing by the shock of the molecular material stored in the icy mantles of the dust.665 Figure 6 displays the J—43 emission integrated between 45 and 62 +. showing a — molecular clump peaking at the position of ECXGOg35.," Figure \ref{hco+} displays the $^{+}$ J=4–3 emission integrated between 45 and 62 , showing a $^{+}$ molecular clump peaking at the position of EGOg35."666"where and f""(x;) is the second derivative of the function f at rj.",where and $f''(x_j)$ is the second derivative of the function $f$ at $x_j$.667" As a consistency check, one can easily show that d?f/dx?=Hf""(vj)+Kf’(aj+1)."," As a consistency check, one can easily show that $d^2f/dx^2=Hf''(x_j)+Kf''(x_{j+1})$."668 The value of the second derivatives are specified by requiring the first derivatives evaluated from the two sides of the grid point to be equal., The value of the second derivatives are specified by requiring the first derivatives evaluated from the two sides of the grid point to be equal.669" Note that this requirement only provides N—2 equations, while there are N second derivatives in total."," Note that this requirement only provides $N-2$ equations, while there are $N$ second derivatives in total."670" The rest of the constraint comes from the boundary conditions on f”(1) and f""(N).", The rest of the constraint comes from the boundary conditions on $f''(1)$ and $f''(N)$.671" In this paper, we simply set [(1)=f""(N)0, which yields the so-calledspline."," In this paper, we simply set $f''(1)=f''(N)=0$, which yields the so-called."672" Finally, we note that the “Log” based interpolation methods are all well defined except when the readouts of some pixels are zero."," Finally, we note that the “Log” based interpolation methods are all well defined except when the readouts of some pixels are zero."673 This is a very rare case in practice due to the presence of noise., This is a very rare case in practice due to the presence of noise.674" However, this situation can in principle exist in simulations."," However, this situation can in principle exist in simulations."675" To cure this problem, one can either change the zeros into tiny positive numbers, or simply avoid interpolating the regions with zeros."," To cure this problem, one can either change the zeros into tiny positive numbers, or simply avoid interpolating the regions with zeros."676" The second option says that if a grid square (regarding the Log-Bilinear and Log-Bicubic methods) or a unitary segment (regarding the Log-Spline method) contains any zero readouts in their four corners or two ends, the finer grid points within them are all set to have zero values."," The second option says that if a grid square (regarding the Log-Bilinear and Log-Bicubic methods) or a unitary segment (regarding the Log-Spline method) contains any zero readouts in their four corners or two ends, the finer grid points within them are all set to have zero values."677 The rest of the grid squares/segments are interpolated independently as usual., The rest of the grid squares/segments are interpolated independently as usual.678" Note that in the Log-Spline method, this means the Spline interpolations are carried out only in those nonzero segments that are isolated by the zeros."," Note that in the Log-Spline method, this means the Spline interpolations are carried out only in those nonzero segments that are isolated by the zeros."679 These two choices usually work similarly well., These two choices usually work similarly well.680" However, when there are extended regions of zero readouts, we find that the second option is better, because it avoids introducing artificial high order fluctuations in the zero regions by methods like Log-Bicubic or Log-Spline."," However, when there are extended regions of zero readouts, we find that the second option is better, because it avoids introducing artificial high order fluctuations in the zero regions by methods like Log-Bicubic or Log-Spline."681where the Hawking temperature is T=rj/zL? and we have defined a critical hardwall temperature T.=ry/7Ll?.,where the Hawking temperature is $T = r_0/\pi L^2$ and we have defined a critical hardwall temperature $T_c = r_{\rm min}/\pi L^2$.682 Η we are willing to tolerate a deviation of the metric from the limiting form given above thenthe expression (48)) is only a good approximation for 0.9<7/7.1., If we are willing to tolerate a deviation of the metric from the limiting form given above thenthe expression \ref{highTeta}) ) is only a good approximation for $0.9 \le T/T_c \le 1$.683" For lower temperatures it is difficult to cleanly separate (he incoming from the outgoing waves,", For lower temperatures it is difficult to cleanly separate the incoming from the outgoing waves.684 The hard wall model is too crucle to access lower temperatures., The hard wall model is too crude to access lower temperatures.685 Nevertheless it does give an indication of how 7/5 might deviate from its conformal value when a confinement scale is introduced., Nevertheless it does give an indication of how $\eta/s$ might deviate from its conformal value when a confinement scale is introduced.686 Note that it does [all below the conjectured lower bound for y/s of 1/4x when T'«1., Note that it does fall below the conjectured lower bound for $\eta/s$ of $1/4\pi$ when $T < T_c$.687 The hard wall result for the shear relaxation time is T 1.49) and T T..(53), The hard wall result for the shear relaxation time is = T T_c and = )^3 2 - )^2 ] )^2 ] - )^2 ] )^2 ] T T_c .688(Larsenetal.2002).. Stephan's Quintet (Ciallagheretal.2001).. and 9922. 2010)..,"\citep{larsen02}, Stephan's Quintet \citep{gallagher01}, and 922 \citep{pellerin}."689 The masses of such complexes vary [rom 10° ML. up to a few 105 M...," The masses of such complexes vary from $10^6\,$ $_\odot$ up to a few $10^8\,$ $_\odot$."690 Bastianetal.(2006) observed star cluster complexes (CC) iu the Antennae with masses of the order cAv105 NL. aud sizes of the order 100 to 200 pc., \cite{bastian06} observed star cluster complexes (CCs) in the Antennae with masses of the order $\approx 10^{6}$ $_{\odot}$ and sizes of the order 100 to 200 pc.691 Peerinetal.(2010) found young massive CC's with tasses between 105 M. and 107 M. and diaimeters between 600 pe aud 1200 pe in the collisional ring galaxy 9922., \cite{pellerin} found young massive CCs with masses between $10^6$ $_\odot$ and $10^{7.5}$ $_\odot$ and diameters between 600 pc and 1200 pc in the collisional ring galaxy 922.692 Meneeletal.(2008 observed icjvidua voung (10 Myr) star clusters associated with cluster complexes iu the Autenae and LIST., \cite{mengel08} observed individual young $\approx$ 10 Myr) star clusters associated with cluster complexes in the Antennae and 1487.693 They compared dyuatmical mass estimates with derived photometric masses and found them i] exceleit agreement. implying that most of them survived the gas removal yhase aid are bound stellar objects.," They compared dynamical mass estimates with derived photometric masses and found them in excellent agreement, implying that most of them survived the gas removal phase and are bound stellar objects."694 These young clusters are sufficiently stable to be used as buildig blocks for numerical similatious., These young clusters are sufficiently stable to be used as building blocks for numerical simulations.695 Bastianetal.(2009) found three 200 to 500 Ayr okd. apparetly stae cltsters in the A1ileuie with very high radial velocities relative to the ealactic disk. iucicating that these star clusers will most Likely become fture halo objects.," \cite{bastian09} found three 200 to 500 Myr old, apparently stable clusters in the Antennae with very high radial velocities relative to the galactic disk, indicating that these star clusters will most likely become future halo objects."696" Oue cluster is surrouded by so far uumerged stela"" eatures in its vicinity.", One cluster is surrounded by so far unmerged stellar features in its vicinity.697 It has alreacy been show. that CCs cau uerge to foruir a variety of sphe‘oldal stellar-cyuamical objects. such as dtra-compac dwarf £aaxies (UCDs). faint [uzzies aud possibly dwaT spheroidal ealaxies (Ixroup:(1998:Fellhauer&Ixroupa2002a.b:Brünsetal.2009).," It has already been shown that CCs can merge to form a variety of spheroidal stellar-dynamical objects, such as ultra-compact dwarf galaxies (UCDs), faint fuzzies and possibly dwarf spheroidal galaxies \citep{krou98,fellhauer02a, fellhauer02b,bruens09}."698. I1 particuleuw. the voung UCD W3 is mos uaturally uiclerstoocl o be a merged massive CC (Felllauer&Iroipa2005).," In particular, the young UCD W3 is most naturally understood to be a merged massive CC \citep{fellhauer05}."699. In the present paper. we apply these ideas to 22019 by analyzug how tl1ο structural parameters of the final merger objects it the outer Galactic halo correlate with the uiderlying CC parameters aud compare them with the observed parameters of 22119.," In the present paper, we apply these ideas to 2419 by analyzing how the structural parameters of the final merger objects in the outer Galactic halo correlate with the underlying CC parameters and compare them with the observed parameters of 2419."700 Iu Section 2.. we describe the metlod amd the parameters used for the calculatious.," In Section \ref{simulations}, we describe the method and the parameters used for the calculations."701 Section κ)+) presents the results of the simulationS. which will be discussed in Section. [..," Section \ref{results} presents the results of the simulations, which will be discussed in Section \ref{discussion}."702 In. Section D> we provide a sumuiLy., In Section \ref{summary} we provide a summary.703 The formation scenario described in this paper starts witl newly born complexes ol star clusters in the Galactic halo., The formation scenario described in this paper starts with newly born complexes of star clusters in the Galactic halo.704 We model the dynamical evolution of various CCs leading to merger objects., We model the dynamical evolution of various CCs leading to merger objects.705 We do not. however. cousider the galaxy-galaxy. interaction. which formed the CCs in the first place.," We do not, however, consider the galaxy-galaxy interaction, which formed the CCs in the first place."706acdvects gas using the piecewise parabolic method. accurate to second-order in space and time. with a Riemann solver at cell boundaries designed to handle strong shock fronts.,"advects gas using the piecewise parabolic method, accurate to second-order in space and time, with a Riemann solver at cell boundaries designed to handle strong shock fronts."707 Our tests ol the FLASII2.5 code and further details about our implementation scheme are detailed in Boss et al. (, Our tests of the FLASH2.5 code and further details about our implementation scheme are detailed in Boss et al. (7082010).,2010).709 Basically. we used the two dimensional. evlindrical coordinate (22.7) version of FLASII2.5. with axisvmmetry about the rotational axis (2).," Basically, we used the two dimensional, cylindrical coordinate $R, Z$ ) version of FLASH2.5, with axisymmetry about the rotational axis $\hat z$ )."710 Multipole sell-gravity was used. including Legendre polynomials up to /=10.," Multipole self-gravity was used, including Legendre polynomials up to $l = 10$."711 The cvlindrical grid was tvpicallv 0.2 pe long in Z ad 0.063 pe wide in A. though in some models the eril was extended to be 0.1 pc long in order to follow the evolution farther downstream.," The cylindrical grid was typically 0.2 pc long in $Z$ and 0.063 pc wide in $R$, though in some models the grid was extended to be 0.4 pc long in order to follow the evolution farther downstream."712 The number of blocks in (Αμ) was 5 in all cases. while the number of blocks in Z (Nyy) was 15 for the grids ancl 20 for the extended exids. with each block consisting of 8x grid points.," The number of blocks in $R$ $N_{BR}$ ) was 5 in all cases, while the number of blocks in $Z$ $N_{BZ}$ ) was 15 for the standard-length grids and 20 for the extended grids, with each block consisting of $8 \times 8$ grid points."713 The number of levels of grid refinement. (NV) was 5 for all models., The number of levels of grid refinement $N_L$ ) was 5 for all models.714 As in Boss οἱ al. (, As in Boss et al. (7152008. 2010). we included compressional heating and radiative cooling. based on the results of Neuleld Ixaufman (1993) for cooling caused by rotational and vibrational transitions of optically thin. warm molecular gas composed of LO. CO. and I.,"2008, 2010), we included compressional heating and radiative cooling, based on the results of Neufeld Kaufman (1993) for cooling caused by rotational and vibrational transitions of optically thin, warm molecular gas composed of $_2$ O, CO, and $_2$."716 As belore. we assumed a radiative cooling rate of Xzz9x10P(T/100)/? erg oE !. where T is the gas temperature in IN and. p is the gas density in ο *.," As before, we assumed a radiative cooling rate of $\Lambda \approx 9 \times 10^{19} (T/100) \rho^2$ erg $^{-3}$ $^{-1}$, where $T$ is the gas temperature in K and $\rho$ is the gas density in g $^{-3}$."717 The gas temperatures were constrained to lie in the range between 10 Ix and 1000 IX. as in Boss et al. (," The gas temperatures were constrained to lie in the range between 10 K and 1000 K, as in Boss et al. ("7182008. 2010). based on the results of Ixaufman Neufeld (1996) for magnetic shock speeds in the desired range of 5 km/sec to 45 km/sec.,"2008, 2010), based on the results of Kaufman Neufeld (1996) for magnetic shock speeds in the desired range of 5 km/sec to 45 km/sec."719 The target dense cloud cores are modeled on Bonner-Ebert (BE) spheres (Bonnor 1956). which are the equilibrium structures for sell-gravitating. isothermal spheres of gas.," The target dense cloud cores are modeled on Bonner-Ebert (BE) spheres (Bonnor 1956), which are the equilibrium structures for self-gravitating, isothermal spheres of gas."720 As in Boss et al. (, As in Boss et al. (7212010). the BE-like spheres are initially isothermal al 10 Ix. with a central clensitw of 124x10 Seem 7I a radius of 0.058 pe. amass of 2.2 AL.. and are stable against collapse [or at least 105 vr.,"2010), the BE-like spheres are initially isothermal at 10 K, with a central density of $1.24 \times 10^{-18}$ g $^{-3}$, a radius of 0.058 pc, a mass of 2.2 $M_\odot$, and are stable against collapse for at least $10^6$ yr."722 The spheres are embedded in an intercloud medium with a density of 3.6x10 σοι and a temperature of 10 IX. Shock waves are launched downward from the top of the grid toward (the spheres (Figure 1) at a speed of 40 km/sec., The spheres are embedded in an intercloud medium with a density of $3.6 \times 10^{-22}$ g $^{-3}$ and a temperature of 10 K. Shock waves are launched downward from the top of the grid toward the spheres (Figure 1) at a speed of 40 km/sec.723 The standard shock front. as used in Boss et al. (," The standard shock front, as used in Boss et al. ("7242008. 2010). has a thickness of 0.003 pe wilh a uniform density of 3.6x107 g em* a mass of 0.015 AL.. a temperature of 1000. IX. and is followed by a post-shock wind with a density of 3.6x10.7? & ! and temperature of 1000 Ix. also moving downward at the same speed as the shock wave.,"2008, 2010), has a thickness of 0.003 pc with a uniform density of $3.6 \times 10^{-20}$ g $^{-3}$, a mass of 0.015 $M_\odot$, a temperature of 1000 K, and is followed by a post-shock wind with a density of $3.6 \times 10^{-22}$ g $^{-3}$ and temperature of 1000 K, also moving downward at the same speed as the shock wave."725 The shock front material is represented by a color field. initially defined to be equal to 1 inside the shock front and 0 elsewhere. which allows the shock wave material to be tracked forward in time (e.g.. Foster Boss 1997).," The shock front material is represented by a color field, initially defined to be equal to 1 inside the shock front and 0 elsewhere, which allows the shock wave material to be tracked forward in time (e.g., Foster Boss 1997)."726production. and max be vital in distinguishing between. example. models of discrete ejections (e.g. van der Laan 66: Alirabel et al.,"production, and may be vital in distinguishing between, for example, models of discrete ejections (e.g. van der Laan 1966; Mirabel et al."727 1998Blandford) or shocks propagating along continuous Lows (e.g. Ixónnigl 1979: Ixaiser. Sunyaev Spruit 2000).," 1998) or shocks propagating along quasi-continuous flows (e.g. Blandford Könnigl 1979; Kaiser, Sunyaev Spruit 2000)."728 We would like to thank the referee. Ralph Spencer. for constructive criticism of this paper. and 1t. Ramachandran for a discussion about radio pulsar rotation measures.," We would like to thank the referee, Ralph Spencer, for constructive criticism of this paper, and R. Ramachandran for a discussion about radio pulsar rotation measures."729 SX.T. is thankful to REBR for support. (grant NOS-02-17577), S.A.T. is thankful to RFBR for support (grant N98-02-17577).730 The Australia Telescope is funded by the Commonwealth of Australia for operation as a National Facility managed by CSIRO., The Australia Telescope is funded by the Commonwealth of Australia for operation as a National Facility managed by CSIRO.731 ΝΤ ASAT results were provided. by. the ASM/INTE teams at MET ancl at the INTIS SOF and GOL at NASA's GSEC., RXTE ASM results were provided by the ASM/RXTE teams at MIT and at the RXTE SOF and GOF at NASA's GSFC.732orientation of the binary orbit in the Galactic reference frame. the systemic kick velocity Vias is assumed to be distributed isotropically in space.,"orientation of the binary orbit in the Galactic reference frame, the systemic kick velocity $\vec{V}_{k,sys}$ is assumed to be distributed isotropically in space."733 The magnitude of Vis is drawn from the velocity distribution derived in the previous step specific to each of the observed systems.," The magnitude of $\vec{V}_{k,sys}$ is drawn from the velocity distribution derived in the previous step specific to each of the observed systems."734 Starting from the randomly generated initial position and velocity. the motion of the DNS ts calculated by numerically integrating the equations of motion from the DNS birth time to the current epoch.," Starting from the randomly generated initial position and velocity, the motion of the DNS is calculated by numerically integrating the equations of motion from the DNS birth time to the current epoch."735 The equations of motion are derived from the Galactic potential of Carlbere Innanen (1987) with updated model parameters of Kuyken Gilmore (1989)., The equations of motion are derived from the Galactic potential of Carlberg Innanen (1987) with updated model parameters of Kuijken Gilmore (1989).736 The DNS birth time is chosen according to the DNS age f; generated in the first step of the analysis., The DNS birth time is chosen according to the DNS age $t_i$ generated in the first step of the analysis.737 Lastly. we use the measured current position of the observed DNS to further constraint the ranges of possible Vj and M»;.," Lastly, we use the measured current position of the observed DNS to further constraint the ranges of possible $V_k$ and $M_{2i}$."738 We set the tolerance of this constraint to be within 100 pe from the observed position., We set the tolerance of this constraint to be within 100 pc from the observed position.739 We test the dependence of our results on this tolerance level by repeating the analysis with a 10 pe. or 50 pe tolerance. but our results do not show any significant changes.," We test the dependence of our results on this tolerance level by repeating the analysis with a 10 pc, or 50 pc tolerance, but our results do not show any significant changes."740 For PSR 3151844904. and PSR J1756-2251. the position constraint is supplemented with the measured current proper motion constraint.," For PSR J1518+4904 and PSR J1756-2251, the position constraint is supplemented with the measured current proper motion constraint."741 For PSR BI534«12. PSR B1913+16. and PSR J0737-3039. the position constraint is supplemented with both the measured proper motion and NSI spin tilt angle constraints.," For PSR B1534+12, PSR B1913+16, and PSR J0737-3039, the position constraint is supplemented with both the measured proper motion and NS1 spin tilt angle constraints."742" The analysis outlined in the previous section allows us to derive PDFs for the kick velocity ¥. NS2’s immediate pre-supernova progenitor mass M»j. the spin tilt 0, of NSI. the systemic radial velocity V4. and the current transverse velocity μι"," The analysis outlined in the previous section allows us to derive PDFs for the kick velocity $V_k$ , NS2's immediate pre-supernova progenitor mass $M_{2i}$, the spin tilt $\theta_t$ of NS1, the systemic radial velocity $V_{rad}$, and the current transverse velocity $V_{trans}$."743 Our primary interest is in V; and M/>;. because they can shed light on the type of supernova that formed NS2. which will be discussed in more detail in the next section.," Our primary interest is in $V_k$ and $M_{2i}$, because they can shed light on the type of supernova that formed NS2, which will be discussed in more detail in the next section."744 The remaining quantities give us some predictions on currently unavailable parameters of the 8 known DNS in our galaxy., The remaining quantities give us some predictions on currently unavailable parameters of the 8 known DNS in our galaxy.745 Instead of focusing on the most likely values as studies have done in the past. we derive confidence levels for the unknown parameters and they are summarized in Table 2.," Instead of focusing on the most likely values as studies have done in the past, we derive confidence levels for the unknown parameters and they are summarized in Table 2."746" Since the V, and M»; constraints are correlated. we present our results in the form of confidence level plots of a 2 dimensional joint. probability distribution of V; - Μαι for each DNS binary. in. addition to the corresponding | dimensional PDF's of and μι."," Since the $V_k$ and $M_{2i}$ constraints are correlated, we present our results in the form of confidence level plots of a 2 dimensional joint probability distribution of $V_k$ - $M_{2i}$, for each DNS binary, in addition to the corresponding 1 dimensional PDF's of $V_k$ and $M_{He}$."747 As pointed out by Willems et al (2006). physical V;parameter constraints are affected by the dimensionality of the PDF used. because of inherent correlations between parameters and projection effects.," As pointed out by Willems et al (2006), physical parameter constraints are affected by the dimensionality of the PDF used, because of inherent correlations between parameters and projection effects."748 This variation is stronger when one solely examines the most likely values (1e.. values at the peak of the PDF) and weakens as broader confidence levels are considered.," This variation is stronger when one solely examines the most likely values (i.e., values at the peak of the PDF) and weakens as broader confidence levels are considered."749 In what follows. the confidence levels of V; and M»; are derived from the 2D joint PDF.," In what follows, the confidence levels of $V_k$ and $M_{2i}$ are derived from the 2D joint PDF."750 To calculate the 2D confidence levels. we first bin the data into a 2D grid and normalize the total probability within the 2D grid to unity.," To calculate the 2D confidence levels, we first bin the data into a 2D grid and normalize the total probability within the 2D grid to unity."751 Then. we add the probability of each bin from the highest to the lowest. until the sum best matches the desired confidence level.," Then, we add the probability of each bin from the highest to the lowest, until the sum best matches the desired confidence level."752" On the other hand. the ranges of Vj. Viens. and 0, are derived from the corresponding 1D PDF."," On the other hand, the ranges of $V_{rad}$, $V_{trans}$, and $\theta_t$ are derived from the corresponding 1D PDF."753 In this case. the confidence level is found by having the shortest range of bins that has a total probability that best matches the desired confidence level.," In this case, the confidence level is found by having the shortest range of bins that has a total probability that best matches the desired confidence level."754 In this binary. the masses of NS1 and NS? are 1.33... and 1.35 M.. respectively.," In this binary, the masses of NS1 and NS2 are 1.33 $M_{\sun}$ and 1.35 $M_{\sun}$ respectively."755" The orbital period 1s P,=0.421 days. with an eccentricity e=0.274."," The orbital period is $P_b = 0.421$ days, with an eccentricity $e = 0.274$ ."756" The angle between NSIs spin axis and the orbital angular momentum axis is 25.07+3.8"" or 155.0?+3.5”.", The angle between NS1's spin axis and the orbital angular momentum axis is $25.0^\circ \pm 3.8^\circ$ or $155.0^\circ \pm 3.8^\circ$.757" The binary is currently located out of the galactic plane. at /=19.8"" and b=48.3"". and 1.02 kpe away from us."," The binary is currently located out of the galactic plane, at $l = 19.8^\circ$ and $b = 48.3^\circ$, and 1.02 kpc away from us."758" The observed proper motion t4,=1.3 mas yr! and Hee.=25.2 mas yr!.", The observed proper motion $\mu_{R.A.} = 1.3$ mas $^{-1}$ and $\mu_{dec.} = 25.2$ mas $^{-1}$.759 At the measured distance this implies VeaΞ64710.32 km s! and Vy...2-1216 kms‘. in the reference frame of the Sun.," At the measured distance this implies $V_{R.A.} = 6.47 \pm 0.32$ km $^{-1}$ and $V_{dec.} = -121 \pm 6$ km $^{-1}$, in the reference frame of the Sun."760 Also. we use the spin down age. which is 210 Myr given by Arzoumanian et al.," Also, we use the spin down age, which is 210 Myr given by Arzoumanian et al."761" 1999, as the upper limit of PSR BI53412 ""s age. instead of the characteristic age."," 1999, as the upper limit of PSR B1534+12 's age, instead of the characteristic age."762 Furthermore. the current orientation of the orbital angular momentum axis is also constrained for this sytem.," Furthermore, the current orientation of the orbital angular momentum axis is also constrained for this sytem."763 The orientation axis can be described by the inclination angle / and the angle of the orbital ascending node on the plane of the sky (Q)., The orientation axis can be described by the inclination angle $i$ and the angle of the orbital ascending node on the plane of the sky $\Omega$ ).764 The sine of the inclination angle is sini = 0.975 (Stairs et al., The sine of the inclination angle is $\sin i$ = 0.975 (Stairs et al.765 2004)., 2004).766 The angle O is 70°+20° or 290°+20°. reckoned north through east.," The angle $\Omega$ is $70^\circ \pm 20^\circ$ or $290^\circ \pm 20^\circ$, reckoned north through east."767 The two solutions correspond to cos;«O0 and cos;>0 respectively (Bogdanov et al., The two solutions correspond to $\cos i < 0$ and $\cos i > 0$ respectively (Bogdanov et al.768 2002)., 2002).769 Since tidal effects between the two neutron stars are negligible. the orbital angular momentum axis keeps a fixed orientation in space.," Since tidal effects between the two neutron stars are negligible, the orbital angular momentum axis keeps a fixed orientation in space."770 From the kinematic analysis of the Galactic motion. the constraints on the present-day orbital inclination and proper motion can therefore be used to determine the systemic kick component V Parallel to the post-supernova orbital angular momentum axis.," From the kinematic analysis of the Galactic motion, the constraints on the present-day orbital inclination and proper motion can therefore be used to determine the systemic kick component $V_{k,sys}^\parallel$ parallel to the post-supernova orbital angular momentum axis."771 Following Kalogera (1996) and Wex et al. (, Following Kalogera (1996) and Wex et al. (7722000). VSYN can also be expressed analytically as where Here. vj (j2 XY.z) Is the x. v. or z component of the natal kick velocity in the frame centered on NS2s progenitor (see figure | in Kalogera for a graphical representation).,"2000), $V_{k,sys}^{\parallel}$ can also be expressed analytically as where Here, $v_{kj}$ $j = x, y, z$ ) is the $x$, $y$, or $z$ component of the natal kick velocity in the frame centered on NS2's progenitor (see figure 1 in Kalogera for a graphical representation)."773" Note that the sign of Vj...| needs to be consistent with the sign of v,-."," Note that the sign of $V_{k,sys}^{\parallel}$ needs to be consistent with the sign of $v_{kz}$."774" Hence. through equation (2). the V,NYN derived from observations gives us a constraint on the y and z components of the natal kick. as well as 0."," Hence, through equation (2), the $V_{k,sys}^{\parallel}$ derived from observations gives us a constraint on the y and z components of the natal kick, as well as $\theta_t$."775 The results are shown in Figures 1 and 2., The results are shown in Figures 1 and 2.776 Right before the second supernova. there must have been RLO from the progenitor of NS? to NSI. as noted previously by Thorsett et al. (," Right before the second supernova, there must have been RLO from the progenitor of NS2 to NS1, as noted previously by Thorsett et al. ("7772005).,2005).778 As seen in the 2D V;-M»; joint. probability distribution in figure |. the kick imparted to NS2 at birth is between 150 and 270 km/s. while the progenitor mass M; is between 1.34 and 3.40 ... both at confidence.," As seen in the 2D $V_k$ $M_{2i}$ joint probability distribution in figure 1, the kick imparted to NS2 at birth is between 150 and 270 km/s, while the progenitor mass $M_{2i}$ is between 1.34 and 3.40 $_\sun$, both at confidence."779 As seen in figure 2. polar kicks (i.e.. 0=0° or 180)are unlikely. and the possible kick directions are asymmetric about the pre- orbital plane (i.e.. 0 2 907).," As seen in figure 2, polar kicks (i.e., $\theta' = 0^\circ$ or $180^\circ$ )are unlikely, and the possible kick directions are asymmetric about the pre-supernova orbital plane (i.e., $\theta'$ = $^\circ$ )."780 This asymmetry comesfrom the constraint on the orientation of the post-supernova orbital angular momentum axis on the sky., This asymmetry comesfrom the constraint on the orientation of the post-supernova orbital angular momentum axis on the sky.781 Moreover. we ürrive at the same conclusion as Thorsett et al. (," Moreover, we arrive at the same conclusion as Thorsett et al. ("782"2005) that 0,~ is very unlikely.",2005) that $\theta_t \sim 155^\circ$ is very unlikely.783 The current radial velocity of this binary, The current radial velocity of this binary784into a disk. leaving a stable (but nonaxisvmmetric) central object.,"into a disk, leaving a stable (but non–axisymmetric) central object."785 In order for this to happen. the star has to have a ratio of rotational kinetic energy to gravitational energv i>09.," In order for this to happen, the star has to have a ratio of rotational kinetic energy to gravitational energy $\beta \geq 0.3$."786 At the end of the collision. our oll-axis collision product has 3~0.045. and is therefore stable against any bar instability.," At the end of the collision, our off-axis collision product has $\beta \sim 0.045$, and is therefore stable against any bar instability."787 However. as the star evolves ancl contracts. it is possible that «7 could becomes high enough to trigger some of these hyvdrodynamic rotational instabilities.," However, as the star evolves and contracts, it is possible that $\beta$ could becomes high enough to trigger some of these hydrodynamic rotational instabilities."788 In this paper. we have limited ourselves to two very specific collisions.," In this paper, we have limited ourselves to two very specific collisions."789 The wo stars involved. in the collision were of equal mass. ancl we only investigated. two choices of impact parameter.," The two stars involved in the collision were of equal mass, and we only investigated two choices of impact parameter."790 “Phere have been other studies. of main-sequence star stellar. collisions which covered more parameter space (Benz&LillsLOST:Lombardiοἱal.1996:Sancdquist.etal. 997).," There have been other studies of main-sequence star stellar collisions which covered more parameter space \cite{BH87,LRS96,SBH97}."791. These studies have clearly and carefully. outlined he properties of collisions. under dilferent. circumstances., These studies have clearly and carefully outlined the properties of collisions under different circumstances.792 These high. resolution simulations were meant to answer some specific questions about. the details of stellar collisions ancl the resulting collision ooducts., These high resolution simulations were meant to answer some specific questions about the details of stellar collisions and the resulting collision products.793 We can ecneralize our current results to all niain-sequence star collisions in elobular clusters under the imitations outlined in the previous papers., We can generalize our current results to all main-sequence star collisions in globular clusters under the limitations outlined in the previous papers.794 Therefore. we eel justificd in claiming that no collision products. have surface convection zones. and that collisions do not. create onglived circumstellar disks.," Therefore, we feel justified in claiming that no collision products have surface convection zones, and that collisions do not create long–lived circumstellar disks."795 The search. to. understand he angular momentum evolution of blue stragelers will iive to turn to other avenues. possibly involving a more detailed look at the combination of stellar evolution. and ivelrodynamic instabilities after the collision.," The search to understand the angular momentum evolution of blue stragglers will have to turn to other avenues, possibly involving a more detailed look at the combination of stellar evolution and hydrodynamic instabilities after the collision."796 This work was supported by PPAUC., This work was supported by PPARC.797 Phe computations reported. here were performed. using the UW Astrophysical Fluids Facility (UI.XEI)., The computations reported here were performed using the UK Astrophysical Fluids Facility (UKAFF).798 DevelopniΗΕ work was performed using the University of Leicester Àathematical Mocelling Contre’s supercomputer which was purchased through the EPSRC strategic equipment initieUive., Development work was performed using the University of Leicester Mathematical Modelling Centre's supercomputer which was purchased through the EPSRC strategic equipment initiative.799 ADD eratelully acknowledges the support of a URE rom the Roval Society. and “PA gratefully acknowledges supxort through a PPARC research studentship.," MBD gratefully acknowledges the support of a URF from the Royal Society, and TA gratefully acknowledges support through a PPARC research studentship."800 The authors would like to thank the referee. Alare Freitag. for his helpful reading. of the nianuscript.," The authors would like to thank the referee, Marc Freitag, for his helpful reading of the manuscript."801 ~3 limes hieher would leave Rossystarbursls (see Figure 5 in section 4.2). still suggesting substantial cust optical depths at short submim wavelengths surpressing high-J CO line emission.,"$\sim $ 3 times higher would leave $\rm R_{65/32}$ (see Figure 5 in section 4.2), still suggesting substantial dust optical depths at short submm wavelengths surpressing high-J CO line emission."802" For à 70Ee21 ""cloaking"" the bulk of the gas and dust in 2220. and conservatively assuming I=L. vields τος25.4 at the emission wavelength of the — fine structure line."," For a $\rm \tau _{850}\ga 1$ “cloaking” the bulk of the gas and dust in 220, and conservatively assuming $\beta $ =1, yields $\rm \tau803_{158}\ga 5.4$ at the emission wavelength of the $^+$ fine structure line."804 This is more (han enough (o almost completely suppress this strong ISAT cooling line into a featureless black body. dust. continuum. favoring large dust optical depths as the cause for its weakness among the other explanations proposed (e.g. Malhotra et al.," This is more than enough to almost completely suppress this strong ISM cooling line into a featureless black body dust continuum, favoring large dust optical depths as the cause for its weakness among the other explanations proposed (e.g. Malhotra et al."805 1997)., 1997).806 Indeed the faint CO J=65 line. besides corroborating the biel dust optical depths at Ht/subnun wavelengths in 2220. also makes them the most likely cause of its line Iuminosity “deficit” (found in ULIRGs. with 2220 having the largest: Luhman et al.," Indeed the faint CO J=6–5 line, besides corroborating the high dust optical depths at IR/submm wavelengths in 220, also makes them the most likely cause of its $^+$ line luminosity “deficit” (found in ULIRGs, with 220 having the largest; Luhman et al."807 1998. 2003).," 1998, 2003)."808 I1 does so by negating a prominent alternative explanation for the line surpression. namely very dense PDRs (where precipitous recombination would remove it from the ISM).," It does so by negating a prominent alternative explanation for the $^+$ line surpression, namely very dense PDRs (where precipitous $^+$ recombination would remove it from the ISM)."809 This is because CO J—65 is a different spectral line (governed by different physies ancl chemistry than } and one that to be Iuminous if (hat alternative explanation held., This is because CO J=6–5 is a different spectral line (governed by different physics and chemistry than $^+$ ) and one that to be luminous if that alternative explanation held.810 Indeed. for dense PDIs immersed in strong Du-UV radiation fields high-J CO lines are expected to be verv luminous and an alternative cooling “channel” to that of the suppressed line. balancing the tremendeous ISM heating expected in ULIRGs (Papadopoulos. Isaak. van der Werl 2007).," Indeed, for dense PDRs immersed in strong far-UV radiation fields high-J CO lines are expected to be very luminous and an alternative cooling “channel” to that of the suppressed $^+$ line, balancing the tremendeous ISM heating expected in ULIRGs (Papadopoulos, Isaak, van der Werf 2007)."811 Finally it must be noted that if high dust optical depths at short subnun wavelengths are responsible for surpressing both the hieh-J CO and the lines in ULIRGs. it follows that the starburst svstems wilh dust-surpressed (high-J)/(1low-J) CO line ratios will also be those with small )/Lig.," Finally it must be noted that if high dust optical depths at short submm wavelengths are responsible for surpressing both the high-J CO and the $^+$ lines in ULIRGs, it follows that the starburst systems with dust-surpressed (high-J)/(low-J) CO line ratios will also be those with small $^+$ $\rm _{IR}$."812" The discovery of dust continuum optical depths that in ULIRGs can remain substantial even out to short submnm wavelengths is the culmination of a series of studies successively ""pushing"" the 721 limit longwards in wavelength for these remarkable svstems (Condon οἱ al 1991: Solomon et al."," The discovery of dust continuum optical depths that in ULIRGs can remain substantial even out to short submm wavelengths is the culmination of a series of studies successively “pushing” the $\tau813_{\lambda }\ga 1$ limit longwards in wavelength for these remarkable systems (Condon et al 1991; Solomon et al."814 1997: Lisenfeld et al., 1997; Lisenfeld et al.815 2000: Sakamoto et al., 2000; Sakamoto et al.816 2003)., 2008).817 Condon et al., Condon et al.818 were the first to point out that hieh dust extinction at far-IR wavelengths is needed to explain radio continuum sizes of ULIRGs that are smaller (han (heir minimum LR emission sizes., were the first to point out that high dust extinction at far-IR wavelengths is needed to explain radio continuum sizes of ULIRGs that are smaller than their minimum IR black-body emission sizes.819 Thev argued (hat re-radiation of the IR lisht from a compact starburst bv dust vlavers” further out that remain optically thick at far-IR wavelengths can vield, They argued that re-radiation of the IR light from a compact starburst by dust “layers” further out that remain optically thick at far-IR wavelengths can yield820(2001).,.821. This teiiplate is based ou the high resolutiou spectrum o: the narrow line Sevtert 1 galaxy PCOO5O|121 (220.06]j. observed with the IHbble Space Telescope.," This template is based on the high resolution spectrum of the narrow line Seyfert 1 galaxy PG0050+124 (z=0.061), observed with the Hubble Space Telescope."822 VestergaarL& Wilkes obtained the emissiou ine template by first fitting and subtracting the power-aw contimmun aud all the absorption/enmission features ron all the clemeuts but Fe., Vestergaard Wilkes obtained the emission line template by first fitting and subtracting the power-law continuum and all the absorption/emission features from all the elements but Fe.823 Afterward an uuuodel was subtracted fro) the residual to obtain a pure template., Afterward an model was subtracted from the residual to obtain a pure template.824 As pointed out. no celission is left iu the 2770-2820 Arrange. due to he Tine subtraction.," As pointed out, no emission is left in the 2770-2820 range, due to the line subtraction."825 We modified the template ollowing bv adding a coustaut fiux deusitv between 2770 and 2820 cequal to the of the mean flux density of the template between 2930 and 2970À., We modified the template following by adding a constant flux density between 2770 and 2820 equal to the of the mean flux density of the template between 2930 and 2970.826. The justification for this operation comes from the theoretical ccmussiou line strength by(2003)., The justification for this operation comes from the theoretical emission line strength by.827. Since the aand the aare not distributed iu the same wav within the BLR. we decided not to fix the Doppler broadening of the ttemplate to the one measured for the cemissiou line.," Since the and the are not distributed in the same way within the BLR, we decided not to fix the Doppler broadening of the template to the one measured for the emission line."828 We have instead rum many tests broadeningi the ttemplate by convolving it with Gaussian profiles with constant FWIIM =7.5.15.22.5A.. corresponding to FWIIM ~930.1860.2800 kms a 2100 rrespectively.," We have instead run many tests broadening the template by convolving it with Gaussian profiles with constant FWHM $=7.5, \ 15, \ 22.5$, corresponding to FWHM $\sim 930, \ 1860, \ 2800$ $/$ s at 2400 respectively."829" A Caussian broadeing with σοςunstaut in wavelcneth leads to slieltly cif&vent velocities over the fitting range. but we have foun| no siguificaito differences in f measured norualization iu ""nuctiou of the three broadening."," A Gaussian broadening with $\sigma$ constant in wavelength leads to slightly different velocities over the fitting range, but we have found no significant differences in the measured normalization in function of the three broadening."830 Caven the typical S/N of our spectra f unonualizatiou is in fact maiilv determined by t fit of the template xoad burps 1ather than individi ffeatures., Given the typical S/N of our spectra the normalization is in fact mainly determined by the fit of the template broad bumps rather than individual features.831 We finally chose to fix FWIIMA15 for the τοι]date., We finally chose to fix $=15$ for the template.832 All the spectra have been shifted to the rest-frame system of reference uxing optical redslüfts., All the spectra have been shifted to the rest-frame system of reference using optical redshifts.833 Even if the optical redshifts are shehtly different from the NIR ones. this will not affect our results since the ppeals wavelength is a free paraiucter in our fitting procedure.," Even if the optical redshifts are slightly different from the NIR ones, this will not affect our results since the peak wavelength is a free parameter in our fitting procedure."834i Moreover. the tteiiplate is constituted by blended multiplets (broadened by the convolution with Cassia profiles with FWIIM =15 Aj). for which the definition of peak wavoleugth is uot straightforward.," Moreover, the template is constituted by blended multiplets (broadened by the convolution with Gaussian profiles with FWHM $=15$ ), for which the definition of peak wavelength is not straightforward."835 We tested a posteriori the difference between the optical redshifts (see Tab., We tested a posteriori the difference between the optical redshifts (see Tab.836 1. and Tab. 2)), \ref{new_source} and Tab. \ref{lit_spec}) )837 and the oues measured from the ppeak wavoleugth (see Tab. D)., and the ones measured from the peak wavelength (see Tab. \ref{tab_mass}) ).838 The average difference is equal to Az=0.02 with a masxiunun of Az=0.08 for JEILE|1217 in this case the Iline profi is severely affected by the atinosplierie absorptions., The average difference is equal to $\Delta$ $=$ 0.02 with a maximum of $\Delta$ $=$ 0.08 for $1411+1217$: in this case the line profile is severely affected by the atmospheric absorptions.839 Our spectra have different wavoleugth coverage because « the different redshifts of the sources aud of the variot iustmruineuts used to collect the data., Our spectra have different wavelength coverage because of the different redshifts of the sources and of the various instruments used to collect the data.840 Siuübulv the sky contanination varies across the sample with redshift., Similarly the sky contamination varies across the sample with redshift.841 For these reasons it is mnupossible to define fixed fitting windows for the eutire sample., For these reasons it is impossible to define fixed fitting windows for the entire sample.842 We focus on the rest-rane region within 2000 and 3500A.. choosing t fittine windows in a wav as homogeneous as possib as a function of the spectral coverage and of the s contanunation.," We focus on the rest-frame region within 2000 and 3500, choosing the fitting windows in a way as homogeneous as possible, as a function of the spectral coverage and of the sky contamination."843 The ft is performed in two steps., The fit is performed in two steps.844 first set of spectral components is given by the sui he power-law contiuuua. the Baler pseudo-conutiuum and the ttemplate.," A first set of spectral components is given by the sum of the power-law continuum, the Balmer pseudo-continuum and the template."845 The free paramcters are the yower-luw slope (0) and its normalization (9= log(Fy)) and the tteuplate normalization (5)., The free parameters are the power-law slope $\alpha$ ) and its normalization $\beta=log(F_0)$ ) and the template normalization $\gamma$ ).846 Since the hnree components are overlapped in this waveleneth ranec. the fit is performed with a \? qinimization ou a suitable erid in the parameter space.," Since the three components are overlapped in this wavelength range, the fit is performed with a $\chi^2$ minimization on a suitable grid in the parameter space."847 Iu this way. we nüuiuize the possibility that our solution represcuts only a local minium of the 4? domain.," In this way, we minimize the possibility that our solution represents only a local minimum of the $\chi^2$ domain."848 The crrors ou the nnormalization are computed by mareimalizing, The errors on the normalization are computed by marginalizing849"irregular, because abrupt and continuous period changes both occur.","irregular, because abrupt and continuous period changes both occur."850 The CCD data shown in Fig., The CCD data shown in Fig.851" 13 confirm the Blazhko effect of V27, which has already been recognized by Oo41."," \ref{v27} confirm the Blazhko effect of V27, which has already been recognized by Oo41."852" Because of crowding problems and irregular fluctuations of the pulsation period, no conclusive result for the modulation period can be drawn from the photographic data."," Because of crowding problems and irregular fluctuations of the pulsation period, no conclusive result for the modulation period can be drawn from the photographic data."853" A separate star, far from the cluster centre."," A separate star, far from the cluster centre."854 The CCD light curve (K00) has a somewhat anomalous shape and small amplitude., The CCD light curve (K00) has a somewhat anomalous shape and small amplitude.855 The anomalous shape of the light curve was already recognised by Oo41., The anomalous shape of the light curve was already recognised by Oo41.856 V29 is the second shortest-period RRab star in M5 (Po—0.4514 d)., V29 is the second shortest-period RRab star in M5 $P_0=0.4514$ d).857 Light curves belonging to the 12 subsets of the photographic observations are shown in Fig. 14.., Light curves belonging to the 12 subsets of the photographic observations are shown in Fig. \ref{v29lc}.858 Variation especially on the upper part of the ascending branch can be suspected., Variation especially on the upper part of the ascending branch can be suspected.859 The pulsation period of V29 decreased by 0.00006 d during the century-long time interval of the observations as shown in Fig., The pulsation period of V29 decreased by 0.00006 d during the century-long time interval of the observations as shown in Fig.860" 1 of Paper I. During the second half of the observations, an ©7000 d fluctuation was superimposed on the gradual period decrease."," 1 of Paper I. During the second half of the observations, an $\approx7000$ d fluctuation was superimposed on the gradual period decrease."861" This is illustrated in the top panel of Fig. 15,"," This is illustrated in the top panel of Fig. \ref{v29},"862 which is a plot of the O—C data from 1946 to 1993., which is a plot of the $O-C$ data from 1946 to 1993.863" Though no significant light-curve variation is detected, the Fourier parameters of the 2-3 year subsets of the photographic data show variations larger than their uncertainties would indicate (Fig. 15))."," Though no significant light-curve variation is detected, the Fourier parameters of the 2–3 year subsets of the photographic data show variations larger than their uncertainties would indicate (Fig. \ref{v29}) )."864" No clear correlation between the variations of the Fourier parameters is evident; however, the changes in the ligh-curve parameters"," No clear correlation between the variations of the Fourier parameters is evident; however, the changes in the ligh-curve parameters"865in gamma-rays.,in gamma-rays.866" Its radio-pulse shape looks much more complicated with something like conal and core component, over a wide range of the pulsar period."," Its radio-pulse shape looks much more complicated with something like conal and core component, over a wide range of the pulsar period."867 The best fit is obtain for an obliquity x=57? and an inclination of ¢=75°., The best fit is obtain for an obliquity $\chi=57^o$ and an inclination of $\zeta=75^o$.868 The gamma-ray light curve adjusts well to Fermi data., The gamma-ray light curve adjusts well to Fermi data.869 Moreover the gamma-ray pulse time lag compared to the middle of the radio pulse matches precisely the measurements., Moreover the gamma-ray pulse time lag compared to the middle of the radio pulse matches precisely the measurements.870" Note that the gamma-ray off pulse emission remains at an appreciable level over the full period, see Fig. 8,,"," Note that the gamma-ray off pulse emission remains at an appreciable level over the full period, see Fig. \ref{fig:FermiCL},"871 on the top right plot., on the top right plot.872" PSR, J0437-4715 is a special case of millisecond pulsar, P=5.76 ms, showing only one gamma-ray pulse combined with a sharp radio pulse."," PSR J0437-4715 is a special case of millisecond pulsar, $P=5.76$ ms, showing only one gamma-ray pulse combined with a sharp radio pulse."873" As explained in a previous discussion, the unique geometry allowing such a behavior needs an obliquity almost equal to the inclination of the line of sight, namely C&x45°."," As explained in a previous discussion, the unique geometry allowing such a behavior needs an obliquity almost equal to the inclination of the line of sight, namely $\zeta \approx \chi \approx 45^o$."874" Indeed, we take x=45° and ¢=40° and arrive at the light curve presented in Fig. 8,,"," Indeed, we take $\chi=45^o$ and $\zeta=40^o$ and arrive at the light curve presented in Fig. \ref{fig:FermiCL},"875" second row, left plot."," second row, left plot."876" For one gamma-ray pulse, our model predicts a delay of 0.5 in phase between radio and gamma-ray pulsars,"," For one gamma-ray pulse, our model predicts a delay of $0.5$ in phase between radio and gamma-ray pulsars,"877"and where and The co-elficients a, and 5, are in general complex functions of the relractive index n and the particle radius to wavelength ratio.",and where and The co-efficients $a_n$ and $b_n$ are in general complex functions of the refractive index $m$ and the particle radius to wavelength ratio.878" For a smooth density distribution. one can write where and 27"" is the associated Legendre function of the first kind."," For a smooth density distribution, one can write where and $P^m_l$ is the associated Legendre function of the first kind."879 substituting eq., Substituting eq.880 7 into eq., \ref{abc1} into eq.881" 2. and integrating over ó. we gel where Considering an axisvmmetry. density distribution with a rotational invariance around sonie axis and using the addition (heorem of spherical harmonie. IN, can be written as"," \ref{abc} and integrating over $\phi$, we get where Considering an axisymmetry density distribution with a rotational invariance around some axis and using the addition theorem of spherical harmonic, $N_{lm}$ can be written as"882Given that the amplitude of the photometric variations was smallest. in the bare (lig. 3)).,"Given that the amplitude of the photometric variations was smallest in the band (Fig. \ref{Fig_deltaft}) ),"883 we estimated the luminosity of PV. Cep [rom the magnitude measured in the low-brightness state on 2008 June LS. assuming that the total [ux in theff ban originated from the photosphere.," we estimated the luminosity of PV Cep from the magnitude measured in the low-brightness state on 2008 June 18, assuming that the total flux in the band originated from the photosphere."884 We adopted a distance of 325 pe for PV. Cop (Straizesctal.1992)... an effective temperature ως=5500 Ix. corresponding to the spectra tvpe GsIO found by Magakian&Alovsessian(2001) 2004)... a total extinction oly=12.0 mag. obtained by dereddening the near-infrared. colour indices (Lorenzettietal.2009) on to theT Tauri locus of the JLf vs. Lfdy. diagram (Mover.Calvet&Lillenbrancl 1997).. and applied the bolomoetric correction tabulated by Llartigan.Strom&(1994) and unreddened colour index V—44=1.69 for the spectral tvpe GS (Ixenvon&Hartmann1995).," We adopted a distance of 325 pc for PV Cep \citep{SCKM}, an effective temperature $T_\mathrm{eff}= 5500$ K, corresponding to the spectral type G8–K0 found by \citet{MM01} , a total extinction $A_\mathrm{V}=12.0$ mag, obtained by dereddening the near-infrared colour indices \citep{Lorenzetti09} on to theT Tauri locus of the $J-H$ vs. $H-K_\mathrm{s}$ diagram \citep*{Meyer97}, and applied the bolometric correction tabulated by \citet*{Hartigan} and unreddened colour index $V-H=1.69$ for the spectral type G8 \citep{KH95}."885".. We obtained £. lvrL.. and a stellar radius #,sz0 ον"," We obtained $L_{*} \approx 17$ $L_{\sun}$, and a stellar radius $R_{*} \approx 4.0$ $_{\sun}$."886 Plotting the results on the WIRD. together with the evolutionary mocels of Palla&Stabler(1999).. results in a mass of MAL; and age of <1 MMwyr.," Plotting the results on the HRD, together with the evolutionary models of \citet{PS99}, results in a mass of $_{\sun}$ and age of $< 1$ Myr."887 We estimated the low-brightness-state disc accretion rate of PV Cep from the luminosity of the 8542 line. measured in 2008 August (Table 5)). using the empirical formula of Dahm(2008).," We estimated the low-brightness-state disc accretion rate of PV Cep from the luminosity of the 8542 line, measured in 2008 August (Table \ref{Tab_lines}) ), using the empirical formula of \citet{Dahm08}."888". We obtain Lisc41Le. and ADSc2.610 ""MAL; /vr."," We obtain $L_\mathrm{acc}^\mathrm{low} \approx 41~L_{\sun}$, and $\dot{M}_\mathrm{acc}^\mathrm{low} \approx 2.6 \times10^{-6}$ $_{\sun}$ /yr."889 The corresponding values for the bright state. estimated. from the flux change of the 8542 line. are about twice higher.," The corresponding values for the bright state, estimated from the flux change of the 8542 line, are about twice higher."890 The results show that the accretion rate of PY Cep is significantly higher than typical PATS accretion rates (~107 ALAL: fer. both in the high- and. low-brightness states. and the accretion Iuminosity is higher than the photospheric Iuminositv.," The results show that the accretion rate of PV Cep is significantly higher than typical PMS accretion rates \citep[$\sim 10^{-8}$ $_{\sun}$ /yr, both in the high- and low-brightness states, and the accretion luminosity is higher than the photospheric luminosity."891Cep between 2004 and 2008 suggest that the fading was associated with decreasing accretion rate., between 2004 and 2008 suggest that the fading was associated with decreasing accretion rate.892 Phe factor of two drop of the accretion Luminosity. derived from the 8542 lines. however. can account. for onlv l mag variation in the optical region.," The factor of two drop of the accretion luminosity, derived from the 8542 lines, however, can account for only $\la~1$ mag variation in the optical region."893 To explain the whole amplituce of d mag. enhanced circumstellar extinction also has to be invoked.," To explain the whole amplitude of $\sim~4$ mag, enhanced circumstellar extinction also has to be invoked."894 The excess extinction could arise from the changes in the inner disc structure., The excess extinction could arise from the changes in the inner disc structure.895 Since Lace27Ly. the modest drop of the accretion rate substantially altered. the total central luminosity. and the decreased:central luminosity made the dust. sublimation radius shrink. that is. a large amount of dust. condensed in the inner disc region during the facing and the subsequent. low-brightness state.," Since $L_\mathrm{acc} > L_*$, the modest drop of the accretion rate substantially altered the total central luminosity, and the decreasedcentral luminosity made the dust sublimation radius shrink, that is, a large amount of dust condensed in the inner disc region during the fading and the subsequent low-brightness state."896" The brieht-state luminosity of Lace|Lyc100L. implies Rou,000.7. AU (οbe.YofDullemond&Alonnier2010).. which might have shrunk to ~O04 AU clue to a [actor of two drop in Lace."," The bright-state luminosity of $L_\mathrm{acc}+L_{*} \approx 100~L_{\sun}$ implies $R_\mathrm{subl}\sim$ 0.7 AU \citep[cf. fig. 7 of ][]{Dullemond}, which might have shrunk to $\sim 0.4$ AU due to a factor of two drop in $L_\mathrm{acc}$."897 Phe observed. drop in the {νο Lux between 2004 ancl 2007 suggests similar decrease in the cust sublimation radius. if we assume that most of this Ες is emitted from the dust. sublimation zone.," The observed drop in the $K_\mathrm{s}$ -band flux between 2004 and 2007 suggests similar decrease in the dust sublimation radius, if we assume that most of this flux is emitted from the dust sublimation zone."898 Assuming normal interstellar eas-to-dust ratio. an excess extinction of ely~3 mag requires a eas column density of ~6.107 7. which. taking into account the size of the dust condensation region. corresponds to a volume density of some 10 .," Assuming normal interstellar gas-to-dust ratio, an excess extinction of $A_\mathrm{V}\sim 3$ mag requires a gas column density of $\sim~6\times10^{21}$ $^{-2}$, which, taking into account the size of the dust condensation region, corresponds to a volume density of some $10^{9}$ $^{-3}$."899 Considering typical midplane gas densities and scaleheights at the dust rim 2010).. the inclination of aand the fact that the disc of PV Cop is at least 10 times more massive than a tvpical ILXe star disc (Llamiclouche 2010).. such densities along the line of sight are likely. and thus the newly formed dusty region may account [or most of the observed optical and near-infrared Lacing of the star.," Considering typical midplane gas densities and scaleheights at the dust rim \citep{Dullemond}, the inclination of and the fact that the disc of PV Cep is at least 10 times more massive than a typical HAe star disc \citep{Hamidouche}, such densities along the line of sight are likely, and thus the newly formed dusty region may account for most of the observed optical and near-infrared fading of the star."900 Phe observed. drop in the mid- ancl Far-infrared. (axes indicates that parts of the optically thin outer IEared disc atmosphere σο into shadow.2," The observed drop in the mid- and far-infrared fluxes indicates that parts of the optically thin outer flared disc atmosphere got into shadow.,"901008. according to the colourmagnitude diagrams. resulted. from variable obscuration along the line of sight.," according to the colour--magnitude diagrams, resulted from variable obscuration along the line of sight."902 A very similar transient. peal both in amplitude and. time-scale appeared in the light curve of PY Cop in 1979. following the end. of the outburst (fig.2ofCohenetal. 1981).. which suggests that the temporary clust-clearing might have been related to the end of the outburst.," A very similar transient peak both in amplitude and time-scale appeared in the light curve of PV Cep in 1979, following the end of the outburst \citep[fig. 2 of][]{Cohen81}, which suggests that the temporary dust-clearing might have been related to the end of the outburst."903 A short-livecl outllow. similar to that observed in V1647 Ori after the end of its outburst in 2006. and probably associated with the rearrangement of the stellar magnetic field in response to the dropped accretion rate (Brittainetal.2010) might have blown out some cust from the line of sight.," A short-lived outflow, similar to that observed in V1647 Ori after the end of its outburst in 2006, and probably associated with the rearrangement of the stellar magnetic field in response to the dropped accretion rate \citep{Brittain10} might have blown out some dust from the line of sight."904 may result from the dust. condensation process., may result from the dust condensation process.905 The duration. of the low-brightness period suggests that the dust. emerging in the line of sight. was not confined to dense clumps or warps in the disc. but rather it was associated with a restructuring of the inner dise due to the changing centralluminosity.," The duration of the low-brightness period suggests that the dust, emerging in the line of sight, was not confined to dense clumps or warps in the disc, but rather it was associated with a restructuring of the inner disc due to the changing centralluminosity."906 max indicate that the dust. condensed. during. the previous vears. started evaporating.," may indicate that the dust, condensed during the previous years, started evaporating."907 The increased κος suggest that invigorated accretion may. play role in the processes., The increased fluxes suggest that invigorated accretion may play role in the processes.908 The time-scales of the large-amplitude photometric luetuations suggest that the inner dust rim is involved., The time-scales of the large-amplitude photometric fluctuations suggest that the inner dust rim is involved.909 Notably. two prominent peaks of the £c light curve (around 22.455.103 and 22.455.143) are separated by," Notably, two prominent peaks of the $I_\mathrm{C}$ light curve (around 2,455,103 and 2,455,143) are separated by"910 Searle&Zinn(1978).. 210 (e.g..Zinn1993:2009).," \citet{searle:78}, $\ga10$ \citep[e.g.,][]{zinn:93,mackey:04,marinfranch:09}."911. (Ibataetal.1994) (DaCosta2003); 2004).," \citep{ibata:94} \citep{dacosta:95,martinezdelgado:02,bellazzini:03}; \citep{martin:04}."912. of sub-groups within the Milky Way globular cluster system reflect the assembly history of the Galactic halo thus remains a eritical unresolved question., of sub-groups within the Milky Way globular cluster system reflect the assembly history of the Galactic halo thus remains a critical unresolved question.913 As the nearest large spiral galaxy. Μο is an attractive alternative target for studying this problem.," As the nearest large spiral galaxy, M31 is an attractive alternative target for studying this problem."914 Its globular cluster system is several times larger than that of the Milky Way. and observational investigation of its halo is less prone to the vagaries of projection and extinction that plague Galactic surveys.," Its globular cluster system is several times larger than that of the Milky Way, and observational investigation of its halo is less prone to the vagaries of projection and extinction that plague Galactic surveys."915 It is known that globular clusters projected near its central regions exhibit some evidence for sub-clustering in position-velocity space that may signal an accretion origin (Ashman&Bird1993:Perrettetal.2003).. although interpretation is difficult because of the complex nature of the inner M31 system.," It is known that globular clusters projected near its central regions exhibit some evidence for sub-clustering in position-velocity space that may signal an accretion origin \citep{ashman:93,perrett:03}, although interpretation is difficult because of the complex nature of the inner M31 system."916 Potentially less confusing are halo regions at projected radii Ry215 kpe. where dynamical times are also longer: however. it is only relatively recently that these remote areas have been targeted by deep wide-field surveys.," Potentially less confusing are halo regions at projected radii $R_{{\rm p}}\ga15$ kpc, where dynamical times are also longer; however, it is only relatively recently that these remote areas have been targeted by deep wide-field surveys."917 Various such studies have shown the M31 halo to be littered with coherent tidal debris features indicative of one or more accretion events (Fergusonetal.2002;Ibata 2007).. and have also facilitated the discovery of significant samples of remote M31 globular clusters (Huxoretal.2008) so that it is now possible to begin assessing how these objects relate to the stellar halo in this galaxy.," Various such studies have shown the M31 halo to be littered with coherent tidal debris features indicative of one or more accretion events \citep{ferguson:02,ibata:07}, and have also facilitated the discovery of significant samples of remote M31 globular clusters \citep{huxor:08} so that it is now possible to begin assessing how these objects relate to the stellar halo in this galaxy."918" Huxoretal.(2010a) have derived the first radial surface-density profile for M31 globular clusters to extend into the far outer halo (Ry,z100 kpe).", \citet{huxor:10} have derived the first radial surface-density profile for M31 globular clusters to extend into the far outer halo $R_{{\rm p}}\approx100$ kpc).919 Their profile exhibits a distinct flattening beyond Ryz30 kpe. very similar to that observed for the metal-poor field halo. and has been interpreted as evidence that accretion processes have played a role in building up both components.," Their profile exhibits a distinct flattening beyond $R_{{\rm p}}\approx30$ kpc, very similar to that observed for the metal-poor field halo, and has been interpreted as evidence that accretion processes have played a role in building up both components."920 In this Letter we use new results from the (PAndAS:MeConnachieetal.2009) to provide the most extensive map to date of globular clusters in the M31 halo. and explore the implications for the origin of this system.," In this Letter we use new results from the \citep[PAndAS;][]{mcconnachie:09} to provide the most extensive map to date of globular clusters in the M31 halo, and explore the implications for the origin of this system."921 PAndAS is an ongoing large program on the Canada-France-Hawait Telescope. utilising the MegaCam imager to obtain a deep panoramic view of M31 and M33.," PAndAS is an ongoing large program on the Canada-France-Hawaii Telescope, utilising the MegaCam imager to obtain a deep panoramic view of M31 and M33."922 First-semester Imaging and data reduction was completed in, First-semester imaging and data reduction was completed in923All clusters have in common that they posses at least several O-type stars with masses in excess of 30MM...,All clusters have in common that they posses at least several O-type stars with masses in excess of $_\odot$.924 The laree number of evolved massive stars. including hypergiants and W-R stars. present in 11 gives evidence that it is the oldest. most evolved cluster in the sample.," The large number of evolved massive stars, including hypergiants and W-R stars, present in 1 gives evidence that it is the oldest, most evolved cluster in the sample."925 Mass segregation seems to be a common feature among young clusters., Mass segregation seems to be a common feature among young clusters.926" As shown in refcluster,fiape.."," As shown in \\ref{cluster_shape}, 1's half-mass radius increases with decreasing stellar mass."927 LI shalf−Wd ≴⋩⋒↙∠∄⋂↙⊉∨∐∖⇆↥∪, A similar trend has been observed by Brandl et \cite{brandl96}) ) for the core radius of the R136 cluster.928 ↗∠∣↾↧∣≀⇂↹⇄∣∖↗∣↰↻∠∕∣∩∏ ∣⋯∣⊓∣∠∣∠∣∣∣∣⋯↗∠∖∣↭∣⊃∣⊃∟↳∣⋯↾⊖∖⋯∥∣≏∐↾∪⋔⊜∣⋯∦⋯≏↧∖∖∣≏↧↳↿∥⋂↑ the ONC and the R136 cluster.," Compared to 3603YC and the Arches cluster, 1 has a rather large half-mass radius of $\approx$ pc, quite similar to the half-mass radii of the ONC and the R136 cluster."929 Since dynamical mass estimates have not yet been derived for any of the clusters. and since the crowding and the presence of bright stars limits the ability to detect the low-mass stellar content of these clusters. mass estimates are m general lower limits.," Since dynamical mass estimates have not yet been derived for any of the clusters, and since the crowding and the presence of bright stars limits the ability to detect the low-mass stellar content of these clusters, mass estimates are in general lower limits."930 11 with a total stellar mass between MM. and MM.. is the most massive. and with a total stellar population of up to 100.000 stars also the most populous among the galactic young massive clusters presented in Table 9..," 1 with a total stellar mass between $_\odot$ and $_\odot$ is the most massive, and with a total stellar population of up to 100,000 stars also the most populous among the galactic young massive clusters presented in Table \ref{sbc}."931" Only the R136 cluster in the 30 Doradus region in the Large Magellanic Cloud. for which Andersen et ((2007)) derive a mass of 2.5 to 3 x10"" MM. counting stars with masses down to MM... Is more massive."," Only the R136 cluster in the 30 Doradus region in the Large Magellanic Cloud, for which Andersen et \cite{andersen07}) ) derive a mass of 2.5 to 3 $ \times 10^4$ $_\odot$ counting stars with masses down to $_\odot$, is more massive."932 We have analysed near-infrared NTT/Sofl observations of the starburst cluster Westerlund 1. which is among the most massive young clusters in the Milky Way.," We have analysed near-infrared NTT/SofI observations of the starburst cluster Westerlund 1, which is among the most massive young clusters in the Milky Way."933" A comparison of colour-magnitude diagrams with theoretical main-sequence and pre-main sequence evolutionary tracks yields. improved extinetion. distance and age estimates of Ay, = 1.134003 mmag. d = 3.5540.17kkpe (DM = 12.7540.10 mmag) and t 2 3.640.7MMyr. respectively."," A comparison of colour-magnitude diagrams with theoretical main-sequence and pre-main sequence evolutionary tracks yields improved extinction, distance and age estimates of $_{\rm Ks}$ = $\pm$ mag, d = $\pm$ kpc (DM = $\pm$ mag) and t = $\pm$ Myr, respectively."934 We derive the slope of the stellar mass function for stars with masses between 3.4 and 27MM..., We derive the slope of the stellar mass function for stars with masses between 3.4 and $_\odot$.935 In an annulus with radit between 0.75 and ppe from the cluster centre. we get a slope of ΓΞ—1.3. tthe Salpeter slope.," In an annulus with radii between 0.75 and pc from the cluster centre, we get a slope of $\Gamma = -1.3$, the Salpeter slope."936 Closer in. the mass function of Westerlund 1 is shallower with Γ=—0.6. while at larger separations from the cluster it is getting steeper. reaching [=-1.6 for separations larger than ppc.," Closer in, the mass function of Westerlund 1 is shallower with $\Gamma = -0.6$, while at larger separations from the cluster it is getting steeper, reaching $\Gamma = -1.6$ for separations larger than pc."937 This ts in good agreement with the change in mass function slope found in the starburst cluster 33603YC (Stolte et citestolte06)), This is in good agreement with the change in mass function slope found in the starburst cluster 3603YC (Stolte et \\cite{stolte06}) ).938 Only considering. stars with masses between 3.0 and 32MM... we derive a half-mass radius of ppc for the cluster.," Only considering stars with masses between 3.0 and $_\odot$, we derive a half-mass radius of pc for the cluster."939 11 exhibits clear deviations from spherical symmetry., 1 exhibits clear deviations from spherical symmetry.940 The an eccentricity of 0.20 with the major axis aligned roughly in the North-South direction., The distribution of stars with masses between 10 and $_\odot$ has an eccentricity of 0.20 with the major axis aligned roughly in the North-South direction.941 The distribution of stars with masses between 3.0 and MM.. is elongated in the same direction with an eccentricity of 0.15., The distribution of stars with masses between 3.0 and $_\odot$ is elongated in the same direction with an eccentricity of 0.15.942 The flattening of the cluster might be explained by rotation of the cluster along its minor axis., The flattening of the cluster might be explained by rotation of the cluster along its minor axis.943 By extrapolation of the observed mass function slopes for different annuli. we derive an upper limit of the total initial stellar cluster mass of =52. 000M...," By extrapolation of the observed mass function slopes for different annuli, we derive an upper limit of the total initial stellar cluster mass of $\approx 52,000$ $_\odot$."944 Stellar evolution at the high mass end should have reduced the total initial cluster mass by «6.000 MM.. over the past MMyr., Stellar evolution at the high mass end should have reduced the total initial cluster mass by $\approx$ $_\odot$ over the past Myr.945 By adding up the individual initial masses of stars directly detected in the cluster. we derived a lower limit for the total present cluster mass of m 2x10! MM...," By adding up the individual initial masses of stars directly detected in the cluster, we derived a lower limit for the total present cluster mass of m $\ge 2\times 10^4$ $_\odot$."946 With a present-day mass of 20.000 to MM. and an initial stellar mass of 52.000 MM... 11 is the most massive starburst cluster identified in the Milky Way to date. and about 10 times as massive as the ONC. and 2 to 4 times as massive a YYC.," With a present-day mass of 20,000 to $_\odot$ and an initial stellar mass of $\approx$ $_\odot$, 1 is the most massive starburst cluster identified in the Milky Way to date, and about 10 times as massive as the ONC, and 2 to 4 times as massive a YC."947 Additional cluster members located outside the field of view of the Sofl data might push the mass of 11 even higher., Additional cluster members located outside the field of view of the SofI data might push the mass of 1 even higher.948 Ina following paper. we will report of the analysis of high-angular resolution adaptive optics data. which trace the pre- sequence population of the cluster down to x0.2 MM...," In a following paper, we will report of the analysis of high-angular resolution adaptive optics data, which trace the pre-main sequence population of the cluster down to $\approx$ $_\odot$."949 Further studies should aim at determining proper motions for individual cluster members as well as more radial velocity measurements., Further studies should aim at determining proper motions for individual cluster members as well as more radial velocity measurements.950 From this precise models of the cluster kinematies and dynamics could be derived. which also might enable one to trace back the dynamical evolution of the cluster since its formation 4 MMyr ago.," From this precise models of the cluster kinematics and dynamics could be derived, which also might enable one to trace back the dynamical evolution of the cluster since its formation $\approx$ Myr ago."951 Il. being the most," 1, being the most"952scale length is close to 2.5Όρο with assumptions on the closure of the hierarchy of velocity moment equations (Cuddeford Amendt. 1992)).,"scale length is close to $2.5953 -3.0 \kpc$ with assumptions on the closure of the hierarchy of velocity moment equations (Cuddeford Amendt, \cite{ca92}) )."954 We must however notice that such analysis using first and second order moments of the distribution neglects available informations like the velocity distribution kurtosis and skewness., We must however notice that such analysis using first and second order moments of the distribution neglects available informations like the velocity distribution kurtosis and skewness.955 The galactic potential can be fitted by a Stücckel potential in the solar neighbourhood., The galactic potential can be fitted by a Stäcckel potential in the solar neighbourhood.956 There the effective potential can be expanded in powers of (η) and z. How well can this expansion be matched by a similar one for a separable potential?, There the effective potential can be expanded in powers of $R_0$ ) and z. How well can this expansion be matched by a similar one for a separable potential?957 For instance. a first order development gives ar and + separable potential which ts then used for the estimate of the force perpendicular to the plane and for the mass density determination in the plane.," For instance, a first order development gives a $r$ and $z$ separable potential which is then used for the estimate of the force perpendicular to the plane and for the mass density determination in the plane."958 Van de Hulst (1962)) and Kent de Zeeuw (1994)) solved the expansion problem with Stücckel potentials., Van de Hulst \cite{vh62}) ) and Kent de Zeeuw \cite{kz91}) ) solved the expansion problem with Stäcckel potentials.959 The expansion includes the third order term +: related to the potential flattening., The expansion includes the third order term $rz^2$ related to the potential flattening.960 If the galactic potential were of Stieckel form we would have an exact measure of its flattening (through the parameter +). the focus of the ellipsoidal coordinate system defined in the Appendix).," If the galactic potential were of Stäcckel form we would have an exact measure of its flattening (through the parameter $z_0$, the focus of the ellipsoidal coordinate system defined in the Appendix)."961 The galactic potential is certainly not exactly of a Stücckel form. but probably not so far from it as long as like here. we consider stars within a small range of excursion from the solar position (3pe radially and 1kpc vertically).," The galactic potential is certainly not exactly of a Stäcckel form, but probably not so far from it as long as like here, we consider stars within a small range of excursion from the solar position $3 \kpc$ radially and $1 \kpc$ vertically)."962 A more general analysis (using orbit computations for instance) could establish accurately the coupling between the vertical and radial motions and its link to any potential., A more general analysis (using orbit computations for instance) could establish accurately the coupling between the vertical and radial motions and its link to any potential.963 The advantage of Stücckel potentials is their tractability and they are certainly sufficient for a first analysis., The advantage of Stäcckel potentials is their tractability and they are certainly sufficient for a first analysis.964 In this paper it allows us to establish formally the importance of the potential flattening on the stellar motions in the solar neighbourhood., In this paper it allows us to establish formally the importance of the potential flattening on the stellar motions in the solar neighbourhood.965 The model is a 3D extension of the 2D distribution functio described in Bienaymé Séechaud (1997))., The model is a 3D extension of the 2D distribution function described in Bienaymé Sécchaud \cite{bs97}) ).966 This model is defined in3D axisymmetric Sticckel potentials where three integrals of motion are known (E. L.. [0).," This model is defined in3D axisymmetric Stäcckel potentials where three integrals of motion are known $E$, $L_z$, $I_3$ )."967 Being a functior of these three integrals. f(E.L..25) is a stationary solutior of the collisionless Boltzmann equation.," Being a function of these three integrals, $f(E,L_z,I_3)$ is a stationary solution of the collisionless Boltzmann equation."968 It is also close to the Schwarzschild distribution for small velocity dispersions anc its associated density is nearly exponential as are the kinematic radial distributions., It is also close to the Schwarzschild distribution for small velocity dispersions and its associated density is nearly exponential as are the kinematic radial distributions.969 The expression of the distribution function is given by (see also the Appendix for more details): is null when L. «0. where 2. =Π(1.}is the radius of circular orbit stars with angular momentum Z.; and having the energy τς].," The expression of the distribution function is given by (see also the Appendix for more details): and is null when $L_z<0$ , where $R_c=R_c(L_z)$ is the radius of circular orbit stars with angular momentum $L_z$ and having the energy $E_{circ}(R_c)$."970" O is the angular velocity. « the epicyelic frequency and in the following. we put R,=RH, and we define the radial dependence of the potential in the galactic plane assuming a power law rotation curve e.(r)i""."," $\Omega$ is the angular velocity, $\kappa$ the epicyclic frequency and in the following, we put $R_{\sigma_z}=R_{\sigma_r}$ and we define the radial dependence of the potential in the galactic plane assuming a power law rotation curve $v_c(r)\sim r^{\alpha}$."971" For a convenientrange of parameters (Fe. Πρ). such a distribution has effectively nearly exponential density and kinematic decreases with respective scale lengths close to Rv and FS, (Fig. 4))."," For a convenientrange of parameters $R_\Sigma$, $R_{\sigma_r}$ ), such a distribution has effectively nearly exponential density and kinematic decreases with respective scale lengths close to $R_\Sigma$ and $R_{\sigma_r}$ (Fig. \ref{fig4}) )."972" Models with extremely large velocity ""Sispersions cannot have small scale lenghts (1.9. large density gradients).", Models with extremely large velocity dispersions cannot have small scale lenghts (i.e. large density gradients).973 In such cases. the effective density scale lengths are different from the input parameter Re and such models o longer have an exponential decrease over a large range of radius (Fig. 4+)).," In such cases, the effective density scale lengths are different from the input parameter $R_\Sigma$ and such models no longer have an exponential decrease over a large range of radius (Fig. \ref{fig4}) )."974 Analysing the Hipparcos data. we will not be concerned by this problem since the velocity dispersions avolved in the data are small.," Analysing the Hipparcos data, we will not be concerned by this problem since the velocity dispersions involved in the data are small."975 The asymmetric drift (AD) relates thevelocity dispersior of any stellar group to its tangential drift., The asymmetric drift (AD) relates thevelocity dispersion of any stellar group to its tangential drift.976 A quite general form is given by Binney Tremaine. 1987 (BT) (Eq.," A quite general form is given by Binney Tremaine, \cite{bt87} (BT) (Eq."977 4.34). valid in the galactic plane at ;=0. relating linearly Vip to the variance o2 (for constant dispersion ratios and kinematic scale length that is independent of stellar types).," 4.34), valid in the galactic plane at $z=0$, relating linearly $V_{AD}$ to the variance $\sigma_r^2$ (for constant dispersion ratios and kinematic scale length that is independent of stellar types)."978 That equation cat be considered as a first order development with a limited range of validity., That equation can be considered as a first order development with a limited range of validity.979 We have computed the asymmetric drift relatior for stationary solutions given by our model and the results are plotted (Fig. 5)), We have computed the asymmetric drift relation for stationary solutions given by our model and the results are plotted (Fig. \ref{fig5}) )980 for different sets of parameters., for different sets of parameters.981 The AD relation is found to be nearly linear in the range of drift from 0 to 30kms.+.," The AD relation is found to be nearly linear in the range of drift from $0$ to $30982 \kms$."983 This interval is apparently the range of validity of the classical equation., This interval is apparently the range of validity of the classical equation.984 Hippareos data as well as other data used to study the asymmetric. drift fall in this range of velocity drift., Hipparcos data as well as other data used to study the asymmetric drift fall in this range of velocity drift.985 However. as seen in. $22.1. the asymmetric. drift relation is not strictly linear.," However, as seen in 2.1, the asymmetric drift relation is not strictly linear."986 Finally for stellar populations having larger drifts. like the thick disc. the use of a linear relation is certainly wrong.," Finally for stellar populations having larger drifts, like the thick disc, the use of a linear relation is certainly wrong."987 The asymmetric drift is neither entirely described by the moment equation 4.34 of Binney Tremaine (1987)) nor by the more exact relations such these plotted on Fig., The asymmetric drift is neither entirely described by the moment equation 4.34 of Binney Tremaine \cite{bt87}) ) nor by the more exact relations such these plotted on Fig.988 5 since. in some range of drift. different modeling may have similar AD relations.," \ref{fig5} since, in some range of drift, different modeling may have similar AD relations."989 Higher moments or marginal velocity distributions help to identify and discriminate between models., Higher moments or marginal velocity distributions help to identify and discriminate between models.990 This is recognized in Fig., This is recognized in Fig.991 6 where various signatures. like drift but also kurtosis and skewness of the ceo distributions. are identifiable.," \ref{fig6} where various signatures, like drift but also kurtosis and skewness of the $v_\theta$ distributions, are identifiable."992 Differences due to a short or a long scale length are visible., Differences due to a short or a long scale length are visible.993" In the case of the shortest scale length 1.8 kpc. with a, increasing. f(eo) stays symmetric andthe maximum shifts regularly."," In the case of the shortest scale length $1.8\kpc$ , with $\sigma_r$ increasing, $f(v_\theta)$ stays symmetric andthe maximum shifts regularly."994 For larger scale length 2.5 kpc. the f(a) distribution becomes highly asymmetric.," For larger scale length $2.5\kpc$ , the $f(v_\theta)$ distribution becomes highly asymmetric."995 At higher scale length 1.5 kpc. the maximum of the co distributions is nearly not shifted.," At higher scale length $4.5\kpc$ , the maximum of the $v_\theta$ distributions is nearly not shifted."996systematics present a challenge to Poisson limited performance.,systematics present a challenge to Poisson limited performance.997" We correct for gain-like variations in high cadence, high precision time series observations in NICMOS, which improve by a factor of two the relative flux measurement."," We correct for gain-like variations in high cadence, high precision time series observations in NICMOS, which improve by a factor of two the relative flux measurement."998" Although the baseline of observations is limited, there is some evidence that the non-repeatability in the G141 grism positioning between orbits affects the system throughput at the 0.1296 level in relative flux."," Although the baseline of observations is limited, there is some evidence that the non-repeatability in the G141 grism positioning between orbits affects the system throughput at the $0.12\%$ level in relative flux."999" When observations taken with similar grism positioning are grouped together, only a few terms related to the PSF positioning and shape are needed to decorrelate the systematics that appear on the HST orbital time scale."," When observations taken with similar grism positioning are grouped together, only a few terms related to the PSF positioning and shape are needed to decorrelate the systematics that appear on the HST orbital time scale."1000 The remaining source of noise is consistent with Poisson and the uncertainty in determining the background level., The remaining source of noise is consistent with Poisson and the uncertainty in determining the background level.1001" Despite the improved modeling of the systematics, the physical transiting planet model contains some degeneracy with the corrections for systematics."," Despite the improved modeling of the systematics, the physical transiting planet model contains some degeneracy with the corrections for systematics."1002" The largest impact is for determining the transit depth, which has an uncertainty 5 times larger than if the Rp/R,,observations were of equivalent quality but free from red-noise systematics."," The largest impact is for determining the transit depth, $R_{p}/R_{\star}$, which has an uncertainty 5 times larger than if the observations were of equivalent quality but free from red-noise systematics."1003" Parameters such as and transit timing are less sensitive to the treatmenta/R, of red-noise systematics.", Parameters such as $a/R_{\star}$ and transit timing are less sensitive to the treatment of red-noise systematics.1004 Southworth(2008) reaches similar conclusions for the analysis of the high quality light curves of HD 209458b taken with the STIS on HST (Brownetal.2001;Knut-sonetal. 2007).," \citet{SOU08} reaches similar conclusions for the analysis of the high quality light curves of HD 209458b taken with the STIS on HST \citep{BRO01B,KNU07}."1005 This emphasizes the necessity of simultaneous fitting of the transit model and systematics in order to provide a more realistic assessment of the uncertainties., This emphasizes the necessity of simultaneous fitting of the transit model and systematics in order to provide a more realistic assessment of the uncertainties.1006" Despite the limitations from needing to correct for the systematics, we greatly improve upon the precision for determining the properties of ddirectly measurable from the light curve such as a/R,, Ryj/R,., px, and transit timing."," Despite the limitations from needing to correct for the systematics, we greatly improve upon the precision for determining the properties of directly measurable from the light curve such as $a/R_{\star}$, $R_{p}/R_{\star}$, $\rho_{\star}$, and transit timing."1007" We find no significant difference in the determination of between the ΝΗΠ and previous determinationsR,/R, in the optical (?,ReproducedinTables1 2)mainBodyRefEnd5934]|TORO08,SOUO08."," We find no significant difference in the determination of $R_{p}/R_{\star}$ between the NIR and previous determinations in the optical \citep[][Reproduced in1008Tables~\ref{tab:planet}~ ."1009" Using the same data set as this study, Tinettietal.(2010) independently find a similar absorption depth averaged over the full wavelength coverage."," Using the same data set as this study, \citet{TIN10} independently find a similar absorption depth averaged over the full wavelength coverage."1010" Using transmission spectroscopy, precise and accurate measurements of the transit depth as a function of wavelength are sensitive to opacity variations in the upper levels (1 mbar) of the planetary atmosphere, and theoretical planetary atmosphere models can be used to interpret these measurements to constrain the abundance of molecules and temperature-pressure profiles (Seager&Sasselov2000;Brown2001;Mad-husudhan&Seager2009;Fortneyetal. 2010)."," Using transmission spectroscopy, precise and accurate measurements of the transit depth as a function of wavelength are sensitive to opacity variations in the upper levels (1 mbar) of the planetary atmosphere, and theoretical planetary atmosphere models can be used to interpret these measurements to constrain the abundance of molecules and temperature-pressure profiles \citep{SEA00,BRO01,MAD09,FOR10}."1011". For example, in their analysis of G141 grism data for HD 149026b, Carteretal.(2009) find a larger absorption depth averaged over the G141 grism when compared to optical and Mid-IR. measurements than the theoretical models predict."," For example, in their analysis of G141 grism data for HD 149026b, \citet{CAR09B} find a larger absorption depth averaged over the G141 grism when compared to optical and Mid-IR measurements than the theoretical models predict."1012 The technical improvements to NICMOS data analysis that we outline in this study may lead to a more precise estimate of the absorption depth in the large archive of NICMOS grism observations of transiting extrasolar planets., The technical improvements to NICMOS data analysis that we outline in this study may lead to a more precise estimate of the absorption depth in the large archive of NICMOS grism observations of transiting extrasolar planets.1013" The transit light curve alone does not allow measuring Ry, and with the quality of light curve from this work, the uncertainty in R, is dominated by the adopted estimate for M, based upon stellar isochrones (Torresetal. 2008)."," The transit light curve alone does not allow measuring $R_{p}$, and with the quality of light curve from this work, the uncertainty in $R_{p}$ is dominated by the adopted estimate for $M_{\star}$ based upon stellar isochrones \citep{TOR08}."1014" Alternatively measuring the parallax ofXO-1, along with Terr, apparent magnitude, and bolometric correction determinations, yields a constraint on R,."," Alternatively, measuring the parallax of, along with $T_{eff}$, apparent magnitude, and bolometric correction determinations, yields a constraint on $R_{\star}$."1015" This constraint on R, and the transit light curve will provide a separate estimate for M,.", This constraint on $R_{\star}$ and the transit light curve will provide a separate estimate for $M_{\star}$.1016" As part of the proposal for these NICMOS observations, Fine Guidance Sensor (FGS) on HST observations were obtained to measure the parallax ofXO-1."," As part of the proposal for these NICMOS observations, Fine Guidance Sensor (FGS) on HST observations were obtained to measure the parallax of."1017". We estimate the FGS distance will provide an estimate of R, with an uncertainty of596,, and this will provide a constraint on M, using the transit light curve from this study."," We estimate the FGS distance will provide an estimate of $R_{\star}$ with an uncertainty of, and this will provide a constraint on $M_{\star}$ using the transit light curve from this study."1018" Asteroseismology provides an additional estimate of p, and has recently been demonstrated to yield very precise estimates for the stellar host and planet in the case of HD 17156b (Nutzman, in preparation), and will be a routine procedure for the transiting planets found around the brighter stars with the Kepler mission (Gillilandetal. 2010)."," Asteroseismology provides an additional estimate of $\rho_{\star}$ and has recently been demonstrated to yield very precise estimates for the stellar host and planet in the case of HD 17156b (Nutzman, in preparation), and will be a routine procedure for the transiting planets found around the brighter stars with the Kepler mission \citep{GIL10}."1019. Measuring the bulk mass and radius of planets with the highest precision will require improvements in understanding the mass and radius of their stellar hosts (Southworth2008)., Measuring the bulk mass and radius of planets with the highest precision will require improvements in understanding the mass and radius of their stellar hosts \citep{SOU08}.1020. The improved precision of the NICMOS time series photometry has provided two additional measurements of the mid-transit time of arrival., The improved precision of the NICMOS time series photometry has provided two additional measurements of the mid-transit time of arrival.1021 The timing between the two HST visits analyzed in this work is different by 2σ from a linear ephemeris., The timing between the two HST visits analyzed in this work is different by $\sigma$ from a linear ephemeris.1022" Also, a single transit measurement from (Cáceresetal.2009) departs most significantly from a linear ephemeris."," Also, a single transit measurement from \citep{CAC09} departs most significantly from a linear ephemeris."1023" Based upon the current measurements of ttransit timings, any bona fide TTV for wwill likely have a peak-to-trough amplitude «90 s, and if the T'TV are concentrated in single events, then single timing precision needs to be «23 s. The HST observations presented in this study achieved o=8.6 s and c—15 s precisions in transit timing."," Based upon the current measurements of transit timings, any bona fide TTV for will likely have a peak-to-trough amplitude $<$ 90 s, and if the TTV are concentrated in single events, then single timing precision needs to be $<$ 23 s. The HST observations presented in this study achieved $\sigma$ =8.6 s and $\sigma$ =15 s precisions in transit timing."1024 The non continuous nature of HST observations negatively impacts the transit timing., The non continuous nature of HST observations negatively impacts the transit timing.1025 The timing precision with HST can be improved by optimizing coverage of the ingress and egress portion of the light curve., The timing precision with HST can be improved by optimizing coverage of the ingress and egress portion of the light curve.1026" However,"," However,"1027" Mi.~-7.4+0.2 Jacobyetal.1992:: Ashmanetal.1995: Harris1999— Ay. Hy, ", $M_V^0 \sim -7.4 \pm 0.2$ \cite{jac92}; \cite{ash95}; \cite{har99} $H_0$ $H_0$ 1028According to current. models. accretion induced. collapse leads to the formation of a rapidlv spinning (a [ew milliseconds) neutron star when an ONeMg white dwarl acereles matter. in a binary system and. reaches the Chandrasekhar limit.,"According to current models, accretion induced collapse leads to the formation of a rapidly spinning (a few milliseconds) neutron star when an ONeMg white dwarf accretes matter in a binary system and reaches the Chandrasekhar limit."1029 Recent calculations have re-alIirmed that an ALC is the expected outcome of thermal time scale mass transfer in such systems with orbital periods of the order of a few days (Ivanova ‘Taam 2004)., Recent calculations have re-affirmed that an AIC is the expected outcome of thermal time scale mass transfer in such systems with orbital periods of the order of a few days (Ivanova Taam 2004).1030 The magnetic lluxes in these cores may thus reflect the magnetic [uxes seen in the isolated white charts., The magnetic fluxes in these cores may thus reflect the magnetic fluxes seen in the isolated white dwarfs.1031 Until recently. it was believed that the magnetic fields of the isolated white chwarls could be described by a single clistribution.," Until recently, it was believed that the magnetic fields of the isolated white dwarfs could be described by a single distribution."1032 However. it is now evident that the distribution is bi-mocdal. comprising of a high and a low field component.," However, it is now evident that the distribution is bi-modal, comprising of a high and a low field component."1033 The high field component (10154 of all white charts) has a clistribution that peaks at logB(C)7.5 with a half width elogB=7.3 (e.g. Wickramasinghe Ferrario 2005)., The high field component $10-15$ of all white dwarfs) has a distribution that peaks at $\log B(G) \sim 7.5 $ with a half width $\sigma\log B = 7.3$ (e.g. Wickramasinghe Ferrario 2005).1034 ‘This distribution declines towards lower fields with very few stars detected in the field range 10° G (magnetic field gap)., This distribution declines towards lower fields with very few stars detected in the field range $10^5 - 10^6$ G (magnetic field gap).1035 The incidence of magnetism rises again towards lower magnetic fields with some 1525% of white clwarls being magnetic at the kilo-Gauss level (Jordan et al., The incidence of magnetism rises again towards lower magnetic fields with some $15 - 25$ of white dwarfs being magnetic at the kilo-Gauss level (Jordan et al.1036 2006)., 2006).1037 Since the new detections of Jordan et al. (, Since the new detections of Jordan et al. (10382006) are at the limit of the sensitivity of current spectropolarimiectric surveys. it appears likely that white dwarls will be found. to he magnetic. with the majority (~ 85%)) belonging to the low field. eroup (~1.000 C.,"2006) are at the limit of the sensitivity of current spectropolarimetric surveys, it appears likely that white dwarfs will be found to be magnetic, with the majority $\sim 85$ ) belonging to the low field group $\lsimeq 1,000$ G)."1039 A field of a kilo-Gauss scales uncer magnetic [ux conservation to a neutron star field of 10° G. Although the peak of the low field. distribution has still to be established observationallv. it is conceivable that the fields will be cistributed in a Caussian manner about a peak that will map on to the observed. field. distribution of the radio MSPs.," A field of a kilo-Gauss scales under magnetic flux conservation to a neutron star field of $10^9$ G. Although the peak of the low field distribution has still to be established observationally, it is conceivable that the fields will be distributed in a Gaussian manner about a peak that will map on to the observed field distribution of the radio MSPs."1040 Given the high mass transfer rates required for ALC. the Ohmic diffusion time scale. will be much larger than the accretion time scale (Cumming 2002). so we expect the white dwarf field to be submerged during the build up of the white dwarf mass prior to collapse.," Given the high mass transfer rates required for AIC, the Ohmic diffusion time scale will be much larger than the accretion time scale (Cumming 2002), so we expect the white dwarf field to be submerged during the build up of the white dwarf mass prior to collapse."1041 For the above scenario to be viable. we need to postulate that the submerged field will re-emerge without decay to its Dux conserved value at the birth of the neutron star.," For the above scenario to be viable, we need to postulate that the submerged field will re-emerge without decay to its flux conserved value at the birth of the neutron star."1042 In this context. we note that there is no evidence of accretion-induced. field. decay in the AAL Lereulis-twpe Cataclysmic Variables. where a highly magnetic white cwarf has been accreting mass over billion vears from a companion.," In this context, we note that there is no evidence of accretion-induced field decay in the AM Herculis-type Cataclysmic Variables, where a highly magnetic white dwarf has been accreting mass over billion years from a companion."1043 In fact. their well studied field configurations are very similar to those observed. and. modelled. in the isolated. high. field magnetic white cwarls (c.g. Wickramasinghe Ferrario 2000 and references therein).," In fact, their well studied field configurations are very similar to those observed and modelled in the isolated high field magnetic white dwarfs (e.g. Wickramasinghe Ferrario 2000 and references therein)."1044 We expect the white dwarf to be spun up to near break up velocity prior to collapse (e.g. like the white cwarts observed. in. dwarf novae)., We expect the white dwarf to be spun up to near break up velocity prior to collapse (e.g. like the white dwarfs observed in dwarf novae).1045 However. angular. momentunm. (and mass) must necessarily be lost during the subsequent collapse to a neutron star (Bailvn Cirinellay 1900) so that detailed mocels are required to establish the expected birth spin and mass distributions of the resulting neutron star.," However, angular momentum (and mass) must necessarily be lost during the subsequent collapse to a neutron star (Bailyn Grindlay 1990) so that detailed models are required to establish the expected birth spin and mass distributions of the resulting neutron star."1046 Dessart et al. (, Dessart et al. (10472006) have conducted: 2.5-climensional racliation-hyclrodvnamics simulations of the ALC of white warfs to neutron stars.,2006) have conducted 2.5-dimensional radiation-hydrodynamics simulations of the AIC of white dwarfs to neutron stars.1048 Their calculations show that these leacl to the formation of neutron stars with rotational periods of a few (2.2-6.3) milliseconcs., Their calculations show that these lead to the formation of neutron stars with rotational periods of a few (2.2-6.3) milliseconds.1049 Hence. even if the binary svstem were to be disrupted. following a kick during the ALC. these “runaway” newly born neutron stars would appear as Isolated. racio-AISPs of the type currently observed.," Hence, even if the binary system were to be disrupted following a kick during the AIC, these “runaway” newly born neutron stars would appear as isolated radio-MSPs of the type currently observed."1050" A proportion of the white dwarfs that could be subjected to ALC will inevitably belong to the hieh field eroup (10?10"" C). and will result. in rapidly rotating (millisecond) pulsars with fields in the range ο10H G on collapse. if we assume magnetic [Lux conservation (see Figure lin FW)."," A proportion of the white dwarfs that could be subjected to AIC will inevitably belong to the high field group $\sim 10^6-10^9$ G), and will result in rapidly rotating (millisecond) pulsars with fields in the range $10^{10}-10^{14}$ G on collapse, if we assume magnetic flux conservation (see Figure 1 in FW)."1051 This proportion could be as high as 50% because highlv magnetic white cwarls tend to be more massive than their non-magnetic (or weakly magnetic) counterparts (mean mass of 0.02M.. Wickramasinghe Ferrario 2005).," This proportion could be as high as $50$ because highly magnetic white dwarfs tend to be more massive than their non-magnetic (or weakly magnetic) counterparts (mean mass of $0.92\Msun$, Wickramasinghe Ferrario 2005)."1052 Assuming that the kicks are not field dependent ancl thus preferentially disrupt. these syvstenis. we may also expect a group of AISPs with high. fields.," Assuming that the kicks are not field dependent and thus preferentially disrupt these systems, we may also expect a group of MSPs with high fields."1053 Llowever. given the much higher spin-down rates of these objects. and their low birth rates as compared to normal racio-pulsars. we expect them to make a small contribution which would be dominated by the lowest field objects in the distribution which would have the longest lifetimes as pulsars.," However, given the much higher spin-down rates of these objects, and their low birth rates as compared to normal radio-pulsars, we expect them to make a small contribution which would be dominated by the lowest field objects in the distribution which would have the longest lifetimes as radio-pulsars."1054" A possible candidate could. be the binary. raclio-pulsar PSR. B0655|64 which has a relatively high magnetic field (B=1.177 C). short orbital period. (2,4,=1.03 d) and is on a nearly circular orbit (Damashek. Llulse 1978: Edwards Bailes 2001)."," A possible candidate could be the binary radio-pulsar PSR B0655+64 which has a relatively high magnetic field $B=1.17^{10}$ G), short orbital period $P_{\rm orb}=1.03$ d) and is on a nearly circular orbit (Damashek, Hulse 1978; Edwards Bailes 2001)."1055 The population svnthesis calculations of Llurley et al. (, The population synthesis calculations of Hurley et al. (10562002) vieldec an ALC rate that is two orders of magnitude higher than the LAINB(CC) rate that results from. the evolution of binary svstems with primaries that are. less massive than 2AL..,2002) yielded an AIC rate that is two orders of magnitude higher than the LMXB(CC) rate that results from the evolution of binary systems with primaries that are less massive than $2\Msun$.1057 X more ctailed investigation of the ALC and core collapse rates and orbital period clistributions expected from such calculations has been presented by Tout et al. (, A more detailed investigation of the AIC and core collapse rates and orbital period distributions expected from such calculations has been presented by Tout et al. (10582007).,2007).1059 Here it is shown that as with the core collapse route.Fypes.," Here it is shown that as with the core collapse route,."1060. We note in particular that a class of long period (Pan=10 d) binary MSPs with He white dwarf companions is predicted. and this closely follows the observed. PuΑπο relationship (Van Ixerkwijk et al.," We note in particular that a class of long period $P_{\rm orb} \ge 10$ d) binary MSPs with He white dwarf companions is predicted, and this closely follows the observed $P_{\rm orb}-M_{WD}$ relationship (Van Kerkwijk et al."1061 2005)., 2005).1062 We conclude this section bv noting that neutron stars hat are formed via ALC may also go. through a mass ransfer phase prior to their switehing-on as radio AISP., We conclude this section by noting that neutron stars that are formed via AIC may also go through a mass transfer phase prior to their switching-on as radio MSP.1063 We herefore expect that the known sample of LAENDs/INNDs will have a contribution. from both neutron stars that jwe resulted. from the core collapse of massive stars and rom the AlC's., We therefore expect that the known sample of LMXBs/IMXBs will have a contribution from both neutron stars that have resulted from the core collapse of massive stars and from the AICs.1064 According to current estimates of the ALC pates. the LAINBs(AIC)/IAINBs(ALC) may dominate over he LAINBs(C'C)/LAINBs(CC).," According to current estimates of the AIC rates, the LMXBs(AIC)/IMXBs(AIC) may dominate over the LMXBs(CC)/IMXBs(CC)."1065 However. because of field submersion. it may be clidifficult at. present to. clistinguish tween these two possibilities.," However, because of field submersion, it may be difficult at present to distinguish between these two possibilities."1066We have studied the preauain-sequeuce evolution of dark stars by following the self-consistent stellar evolution.,We have studied the pre-main-sequence evolution of dark stars by following the self-consistent stellar evolution.1067 We have suitably modified the code to incorporate the energy ecucration from spherically distributed DM., We have suitably modified the code to incorporate the energy generation from spherically distributed DM.1068 Oi base mode with my=100GeV aud dALidT=LON Mvis| shows the characteristic features of the dark star phase: he large DAL annihilation cnerey expands the star. qnakine the star to be iu gravitational equilibrimu.," Our base model with $m_{\chi}= 100 \ \mathrm{GeV}$ and $dM/dT=1.0 \times $ $ \ \mathrm{M_{\odot} \ yrs^{-1}}$ shows the characteristic features of the dark star phase; the large DM annihilation energy expands the star, making the star to be in gravitational equilibrium."1069 The lower temperature is one of he peculiar properties of the dark star., The lower temperature is one of the peculiar properties of the dark star.1070 This stable phase coutinues until the cucreyv supply TOM DAI aunihilation becomes insuffiicicnt to naintain the stable structure., This stable phase continues until the energy supply from DM annihilation becomes insufficient to maintain the stable structure.1071 Finally. the star collapses rapidly iux reaches the main-sequence ohase.," Finally, the star collapses rapidly and reaches the main-sequence phase."1072 At this point. the stellar mass has grown up to AS—9001000NL...," At this point, the stellar mass has grown up to $M \sim 900 - 1000 \ \mathrm{M_{\odot}}$."1073 Such features in dark star phase are all consistent with the findings of oxevious works of Toccoetal.(2008):Spolvareal.(2009) who curploved imch simpler stellar nodels.," Such features in dark star phase are all consistent with the findings of previous works of \citet{iocco08, spolyar09}1074 who employed much simpler stellar models."1075 The dark star mocel has effectively two parameuS., The dark star model has effectively two parameters.1076 One of thei is the eas mass accretion rate which deteriuues the evolution of the gas aud DA distribution., One of them is the gas mass accretion rate which determines the evolution of the gas and DM distribution.1077 Cosimological simulations predic a variety of gas accretion rates;, Cosmological simulations predict a variety of gas accretion rates.1078 For a smal accretion rate. the period of the dark star phase lucreases whereas the final stellar mass decreases oaun?IOADL.," For a small accretion rate, the period of the dark star phase increases whereas the final stellar mass decreases $M \sim 500 \ \mathrm{M_{\odot}}$."1079 Note however that the mass is still lareer than the standard Pop IIT (no-DM) case (LO00200ALL)., Note however that the mass is still larger than the standard Pop III (no-DM) case $100 - 200\ \mathrm{M_{\odot}})$.1080 Another parameter is the DAL particle mass which determines the energy eeneration rate., Another parameter is the DM particle mass which determines the energy generation rate.1081 For a small DAL particle mass. the period of dark star phase becomes longer aud the final stellar mass becomes larger (sce Table 5)).," For a small DM particle mass, the period of dark star phase becomes longer and the final stellar mass becomes larger (see Table \ref{table5}) )."1082 All models pass through the dark star phase and this phase is maintained by a little DM fuel which is less than 0.1% of the stellar mass., All models pass through the dark star phase and this phase is maintained by a little DM fuel which is less than $0.1 \ \%$ of the stellar mass.1083 If the first stars in the universe have undergone such phase. there are exotic stars which are cool aud luassive m the early universe.," If the first stars in the universe have undergone such phase, there are exotic stars which are cool and massive in the early universe."1084 Formation of very iiassive dark stars has i miportaut duplication., Formation of very massive dark stars has an important implication.1085 Such stars eveutually collapse gravitationally. to form massive black holes with masses as large as 1000AL...," Such stars eventually collapse gravitationally, to form massive black holes with masses as large as $1000 \ M_{\odot}$."1086 The remnant black holes can also erow by accretion or by mergers to seed super-anassive black holes., The remnant black holes can also grow by accretion or by mergers to seed super-massive black holes.1087 The exotic feature of dark stars appearance. Iluniuous and cool. may be able to use as the powerful clue to search such stars;," The exotic feature of dark star's appearance, luminous and cool, may be able to use as the powerful clue to search such stars."1088 Our calculations show that first stars can grow to be dark stars with huuimositv —few10°L. at most and they will not be detectable hyJWST., Our calculations show that first stars can grow to be dark stars with luminosity $\sim \mathrm{few} \ \ 10^7 \ \mathrm{L_{\odot}}$ at most and they will not be detectable by.1089 However. Freeseetal.(2010) estimate for and such very iassive and bright stars (about 109ΤΕΕς} can be detected by JAVST.," However, \citet{freese10} estimate for and such very massive and bright stars (about $10^{9-11} \ L_{\odot}$ ) can be detected by JWST."1090 Sanclicket argue that dark star remmauts might survive to the present day in the Milkv Was. leaving οταν signatures frou DAL aunihlilatio-," \citet{sandick10} argue that dark star remnants might survive to the present day in the Milky Way, leaving $\gamma$ -ray signatures from DM annihilation."1091 Tn future work. we will include the nuclear reaction to caleulate the first star evolution from the dark star phase to the main-sequence phase conipletelv.," In future work, we will include the nuclear reaction to calculate the first star evolution from the dark star phase to the main-sequence phase completely."1092 It would be also interesting to include capture of DM particles., It would be also interesting to include capture of DM particles.1093 Ouce the stars have contracted aud stop erowing in mass. there will be uo more DM supplied by adiabatic contraction.," Once the stars have contracted and stop growing in mass, there will be no more DM supplied by adiabatic contraction."1094 However. DM particle may still be captured by the star.," However, DM particle may still be captured by the star."1095 Previous studies showed that this process cau make the star back iu the dark star phase again., Previous studies showed that this process can make the star back in the dark star phase again.1096 We will explore the effect of DM capture aud make a complete evolutionary model of the first stars with DAL annihilation., We will explore the effect of DM capture and make a complete evolutionary model of the first stars with DM annihilation.1097 This work was supported bv the Crauts-iu-Add foy Scientific Research (20011005. 20105001) from the MENT of Japan aud the Cuaauts-iu-Aid for Young Scicutists (9) 20671003 bv the Japan Society for the Promotion of Science. and support from World Premicr LIuteruational Research Center Initiative (PI Initiative). MEXT. Japan.," This work was supported by the Grants-in-Aid for Scientific Research (20041005, 20105004) from the MEXT of Japan and the Grants-in-Aid for Young Scientists (S) 20674003 by the Japan Society for the Promotion of Science, and support from World Premier International Research Center Initiative (WPI Initiative), MEXT, Japan."1098shows the constraints on these parameters.,shows the constraints on these parameters.1099 In contrast. the 1991 observation is very poorly fit with this model (reduced 6T with 4 clot).," In contrast, the 1991 observation is very poorly fit with this model (reduced $\chi^2$ =6.7 with 4 d.o.f.)."1100 The unabsorbed kkeW flux for the fit to the 1993 spectrum is 910“erestem27.," The unabsorbed keV flux for the fit to the 1993 spectrum is $9\times10^{-12}\rm\,erg\,s^{-1}\,cm^{-2}$."1101 Including a bolometric correction factor of 2.4. this corresponds to a luminosity of G6.10eres tat the LIIPPARCOS distance of 4 Draconis ppe: ?2)).," Including a bolometric correction factor of 2.4, this corresponds to a luminosity of $6\times10^{31}\rm\,erg\,s^{-1}$ at the HIPPARCOS distance of 4 Draconis pc; \ncite{Perryman97}) )."1102 The fit residuals for the 1991 spectrum show a minimum. around kkeV reffig-spec)) which corresponds to the maximum. of the cHective area of the PSPC ancl so must represent a rue minimum in the X-rav Bux., The fit residuals for the 1991 spectrum show a minimum around keV \\ref{fig-spec}) ) which corresponds to the maximum of the effective area of the PSPC and so must represent a true minimum in the X-ray flux.1103 No physically-plausible »ure-enmission model can reproduce this minimum. but a xutiallv-ionised. absorber does so naturally.," No physically-plausible pure-emission model can reproduce this minimum, but a partially-ionised absorber does so naturally."1104 Photoelectric absorption by cold. cosmic-abundance material increases to ow energies. but soft. photons can leak through if. low enerev edges have been removed. through ionisation.," Photoelectric absorption by cold cosmic-abundance material increases to low energies, but soft photons can leak through if low energy edges have been removed through ionisation."1105 We ind that such a model readily reproduces the observed. 1991 spectrum. and we plot a tvpical fit in reffig-absori (with. hydrogen column density of 4 and ionisation parameter £—5.8).," We find that such a model readily reproduces the observed 1991 spectrum, and we plot a typical fit in \\ref{fig-absori} (with hydrogen column density of $4\times10^{23}\rm\,cm^{-2}$ and ionisation parameter $\xi$ =5.8)."1106 Unfortunately the combination of low spectral resolution and low signal-to-noise prevents a unique fit to an ionised absorption moclel and we cannot well constrain the properties of the absorbing medium., Unfortunately the combination of low spectral resolution and low signal-to-noise prevents a unique fit to an ionised absorption model and we cannot well constrain the properties of the absorbing medium.1107 However. the wind of the red giant star is an obvious candidate absorber and the X-ray source itself may supply the photo-ionising flux.," However, the wind of the red giant star is an obvious candidate absorber and the X-ray source itself may supply the photo-ionising flux."1108 The dillerence between the 1991 and 1993 observations may be explained. in part or entirely by a changing column density and/or ionisation fraction along our line of sight through the wind., The difference between the 1991 and 1993 observations may be explained in part or entirely by a changing column density and/or ionisation fraction along our line of sight through the wind.1109 The lighteurves from all three ROSAT observations reveal variability on short. timescales., The lightcurves from all three ROSAT observations reveal variability on short timescales.1110 The lighteurve. [rom the kks LRD observation is presented in reffig-Ic.., The lightcurve from the ks HRI observation is presented in \\ref{fig-lc}.1111 Lt shows strong variability on timescales between minutes and days; but there is no evidence for the four-hour periodic modulation claimed by ο from IUIS observations.," It shows strong variability on timescales between minutes and days, but there is no evidence for the four-hour periodic modulation claimed by \scite{Reimers88} from IUE observations."1112 reffie-pspe shows the power spectrum of the LEE lightcurve. in which no obvious periodic signal is apparent.," \\ref{fig-pspe} shows the power spectrum of the HRI lightcurve, in which no obvious periodic signal is apparent."1113 There is excess power close το the ROSAT orbital period (OGmmin). with the strongest peak at a slightly shorter period mmin). and there is another suggestive peak at mmin. but neither are sulliciently strong to represent a conclusive detection of periodic modulation.," There is excess power close to the ROSAT orbital period min), with the strongest peak at a slightly shorter period min), and there is another suggestive peak at min, but neither are sufficiently strong to represent a conclusive detection of periodic modulation."1114"yyields Γ=1.99+0.15 (X2,,=0.89, 21 dof, Table 4)).","yields $\Gamma = 1.99 \pm 0.15$ $\chi^2_{red}=0.89$, 21 dof, Table \ref{tb:fit}) )."1115" Fitting with ffree results in a slightly flatter power-law slope (I'=1.92+0.14, x2,4—0.87, 20 dof) and very low ssuggesting the presence of a soft excess."," Fitting with free results in a slightly flatter power-law slope $\Gamma = 1.92 \pm 0.14$, $\chi^2_{red}=0.87$, 20 dof) and very low suggesting the presence of a soft excess."1116 However the signal-to-noise is insufficient to distinguish between models for a soft excess and the power-law slope is typical of radio-quiet quasars (?) so no more complex modeling was carried out.," However the signal-to-noise is insufficient to distinguish between models for a soft excess and the power-law slope is typical of radio-quiet quasars \citep{2005A&A...432...15P}1117 so no more complex modeling was carried out."1118" Fluxes were determined using the initial fit, assuming GalacticNu."," Fluxes were determined using the initial fit, assuming Galactic."1119. There are additional X-ray counts around the quasar core which are not associated with the envelopes of radio emission., There are additional X-ray counts around the quasar core which are not associated with the envelopes of radio emission.1120 The emission occurs on both sides of the quasar but the number of counts is too low to constrain its spatial distribution., The emission occurs on both sides of the quasar but the number of counts is too low to constrain its spatial distribution.1121" We extracted counts in the broad band (0.3—8 keV) from an annular region centered on 2270.1 and extending from 2""—7.""5 (~64 kpc) excluding two segments (opening angles: North:60°, South: 40°)) containing the radio-related regions, this excluded of the annulus."," We extracted counts in the broad band $-$ 8 keV) from an annular region centered on 270.1 and extending from $2\arcsec-7.\arcsec5$ $\sim$ 64 kpc) excluding two segments (opening angles: North:, South: ) containing the radio-related regions, this excluded of the annulus."1122" Background counts (34) were estimated from an annular region extending from 10""—20"".", Background counts (34) were estimated from an annular region extending from $10''-20''$.1123" Twenty-seven counts were detected where 4.3+0.7 are expected, a detection which is highly significant (Table 3))."," Twenty-seven counts were detected where $\pm$ 0.7 are expected, a detection which is highly significant (Table \ref{tb:counts}) )."1124 The net extended counts are, The net extended counts are1125to separations comparable to the distance and velocity errors of the data.,to separations comparable to the distance and velocity errors of the data.1126" In we show the range of measured for each of the six simulations described in4,, overplotting the data shown in3."," In we show the range of measured for each of the six simulations described in, overplotting the data shown in."1127". As described in these results correspond to mock BHB catalogues for 500 Solar observers located in the *Galactic plane', with typical observation errors in distance and velocity."," As described in these results correspond to mock BHB catalogues for 500 Solar observers located in the `Galactic plane', with typical observation errors in distance and velocity."1128 The dashed and solid lines bounding each coloured region in correspond to the 10th and 90th percentile values we obtain in each bin ofA)., The dashed and solid lines bounding each coloured region in correspond to the 10th and 90th percentile values we obtain in each bin of.1129". At a given scale in our A metric, the strength of the clustering signal varies considerably from halo to halo and between individual observers."," At a given scale in our $\Delta$ metric, the strength of the clustering signal varies considerably from halo to halo and between individual observers."1130" The overall trend of is similar to the observations in all haloes, although the clustering signal rises more steeply on small scales in most of the simulations."," The overall trend of is similar to the observations in all haloes, although the clustering signal rises more steeply on small scales in most of the simulations."1131" On scales A<4 kpc, significant clustering is detected by all observers in five of the six haloes."," On scales $\Delta < 4$ kpc, significant clustering is detected by all observers in five of the six haloes."1132 The exception is halo Aq-E (dark blue)., The exception is halo Aq-E (dark blue).1133" This stellar halo is highly concentrated, and at r>20 kpc is dominated by a single radial stream (see figures 6 and 7 of Cooper 2010)."," This stellar halo is highly concentrated, and at $r>20$ kpc is dominated by a single radial stream (see figures 6 and 7 of Cooper 2010)."1134" We find that two haloes, Aq-E and Aq-F (red), are consistent with the observed on all scales."," We find that two haloes, Aq-E and Aq-F (red), are consistent with the observed on all scales."1135 The structure of Aq-F is atypical for the sample — most of its stars are accreted in a late 3:1 merger and its surface brightness at the Solar radius is substantially higher than current estimates for the Milky Way halo., The structure of Aq-F is atypical for the sample – most of its stars are accreted in a late 3:1 merger and its surface brightness at the Solar radius is substantially higher than current estimates for the Milky Way halo.1136" In projection, Aq-F resembles the ‘shell’-dominated haloes observed in a number of nearby elliptical galaxies."," In projection, Aq-F resembles the `shell'-dominated haloes observed in a number of nearby elliptical galaxies."1137" Meanwhile haloes Aq-A (black), Aq-B (cyan) and Aq-D (green) are marginally inconsistent with the data: below A~4 kpc, ~90 per cent of mock observations in these haloes imply a greater degree of clustering than we find for the Milky Way, particularly on small scales."," Meanwhile haloes Aq-A (black), Aq-B (cyan) and Aq-D (green) are marginally inconsistent with the data: below $\Delta\sim4$ kpc, $\sim90$ per cent of mock observations in these haloes imply a greater degree of clustering than we find for the Milky Way, particularly on small scales."1138" Aq-C (purple) is entirely inconsistent with the Milky Way observations on all scales, showing a much higher degree of clustering."," Aq-C (purple) is entirely inconsistent with the Milky Way observations on all scales, showing a much higher degree of clustering."1139" Beyond 20 kpc, the sky of an observer in Aq-C is dominated by two bright tidal streams on wide (~100 kpc) orbits."," Beyond $20$ kpc, the sky of an observer in Aq-C is dominated by two bright tidal streams on wide $\sim100$ kpc) orbits."1140" Although their orbital planes are approximately coincident with our definition of the Galactic plane, nevertheless sections of these streams intrude on the SDSS footprint at low Galactic latitudes."," Although their orbital planes are approximately coincident with our definition of the Galactic plane, nevertheless sections of these streams intrude on the SDSS footprint at low Galactic latitudes."1141 The DR6 footprint and the cut on extra-planar height in the Xue (2008) sample exclude stars near the Galactic plane from our clustering analysis., The DR6 footprint and the cut on extra-planar height in the Xue (2008) sample exclude stars near the Galactic plane from our clustering analysis.1142 illustrates how our definition of the Galactic plane influences the halo clustering signal., illustrates how our definition of the Galactic plane influences the halo clustering signal.1143" In panel (a) the orientation of the Galactic plane with respect to the halo is chosen randomly for each of the 500 mock observers (i.e. observers are distributed over a sphere of radius r;;=8 kpc), whereas in panel (b) the galactic Z direction is aligned with the minor axis of the halo for all observers as in4."," In panel (a) the orientation of the Galactic plane with respect to the halo is chosen randomly for each of the 500 mock observers (i.e. observers are distributed over a sphere of radius $r_{\sun}=8$ kpc), whereas in panel (b) the galactic Z direction is aligned with the minor axis of the halo for all observers as in."1144". To focus on the effects of this alignment, the distances and velocities of stars in these two panels have been convolved with observational errors."," To focus on the effects of this alignment, the distances and velocities of stars in these two panels have been convolved with observational errors."1145 The systematically higher clustering signals in panel (b) of suggest that the plane perpendicular to the minor axis of the dark matter halo is special., The systematically higher clustering signals in panel (b) of suggest that the plane perpendicular to the minor axis of the dark matter halo is special.1146" In Cooper (2010) and above, we have noted the strong correlation between the shape of the dark halo and the inner regions of the stellar halo."," In Cooper (2010) and above, we have noted the strong correlation between the shape of the dark halo and the inner regions of the stellar halo."1147" This alignment of halo structure also extends, more loosely, to other prominent stellar halo structures at large distances."," This alignment of halo structure also extends, more loosely, to other prominent stellar halo structures at large distances."1148 An overall flattening of the stellar halo arises because our dark matter haloes are embedded in long-lived filaments of the cosmic web., An overall flattening of the stellar halo arises because our dark matter haloes are embedded in long-lived filaments of the cosmic web.1149" Typically one or two such filaments dominate the infall directions of both satellite galaxies and smoothly accreted dark matter, which also contributes to the shape of the dark halo (e.g. Libeskind 2005; Lovell 2011; Wangal. 2011; Vera-Ciroal. 2011)."," Typically one or two such filaments dominate the infall directions of both satellite galaxies and smoothly accreted dark matter, which also contributes to the shape of the dark halo (e.g. Libeskind 2005; Lovell 2011; Wang 2011; Vera-Ciro 2011)."1150 The distribution of stars stripped from infalling satellites echoes the large-scale correlation of their orbital planes., The distribution of stars stripped from infalling satellites echoes the large-scale correlation of their orbital planes.1151" Because of this flattened global structure, the distribution of halo stars in our choice of Galactic plane tend to be more smoothly distributed (i.e. this plane contains more diffuse phase-mixed material as well as coherent substructure)."," Because of this flattened global structure, the distribution of halo stars in our choice of Galactic plane tend to be more smoothly distributed (i.e. this plane contains more diffuse phase-mixed material as well as coherent substructure)."1152 Panel (b) demonstrates how the ‘contrast’ of small scale substructure in the outer halo is enhanced when these smoother components are excluded from the clustering analysis (through a combination of the SDSS footprint and the cut on |Ζ])., Panel (b) demonstrates how the `contrast' of small scale substructure in the outer halo is enhanced when these smoother components are excluded from the clustering analysis (through a combination of the SDSS footprint and the cut on $|Z|$ ).1153" This is particularly true in the case of Aq-F, where the majority of the mass in the halo is contributed by one extensive and relatively ‘smooth’ component."," This is particularly true in the case of Aq-F, where the majority of the mass in the halo is contributed by one extensive and relatively `smooth' component."1154" By contrast, in Aq-C the average clustering amplitude on large scales when we fix the Galactic plane."," By contrast, in Aq-C the average clustering amplitude on large scales when we fix the Galactic plane."1155" As noted above, in this case the massive coherent streams that dominate the clustering signal of this halo mostly fall outside the SDSS footprint."," As noted above, in this case the massive coherent streams that dominate the clustering signal of this halo mostly fall outside the SDSS footprint."1156 Panel (c) of shows the randomly aligned observations of panel (a) convolved with observational errors in distance and velocity., Panel (c) of shows the randomly aligned observations of panel (a) convolved with observational errors in distance and velocity.1157" These errors ‘smooth out’ the halo, suppress the clustering signal overall and increase the variance between observers on small scales."," These errors `smooth out' the halo, suppress the clustering signal overall and increase the variance between observers on small scales."1158" Again the effect is most pronounced for Aq-F, where blurring of the dominant smooth component further decreases the contrast of substructure."," Again the effect is most pronounced for Aq-F, where blurring of the dominant smooth component further decreases the contrast of substructure."1159 In most cases these two effects (alignment and observational errors) counteract each other to produce the distribution of signals shown in4.., In most cases these two effects (alignment and observational errors) counteract each other to produce the distribution of signals shown in.1160" In the case of Aq-E the signal suffers disproportionately from errors in the aligned configuration, perhaps because this signal is due to a small number of pairs at large distances."," In the case of Aq-E the signal suffers disproportionately from errors in the aligned configuration, perhaps because this signal is due to a small number of pairs at large distances."1161" Finally, in we examine differences between nearby and"," Finally, in we examine differences between nearby and"1162velocity dispersion. the more the laver resists gravitational collapse.,"velocity dispersion, the more the layer resists gravitational collapse."1163 In an LL laver (P= IxXIx. ppz mmg ). the sound speed is c +. while it can be as low as & in an Hl» laver (P = 20KK. ppc μπαμ) (AleCrayv Walatos 1987).," In an I layer (T $\simeq$ K, $\mu ~\simeq$ $_H$ ), the sound speed is $\simeq$ $^{-1}$, while it can be as low as $\simeq$ $^{-1}$ in an $_2$ layer (T $\simeq$ K, $\mu ~\simeq$ $_H$ ) (McCray Kafatos 1987)."1164 However. the turbulence and magnetic fields of the shell will increase these values.," However, the turbulence and magnetic fields of the shell will increase these values."1165" To some extent. they can be represented. by an additional pressure term in the expression of ὃς. i.e. where D, is the magnetic field in the shell anc 7, is the contribution of turbulence."," To some extent, they can be represented by an additional pressure term in the expression of $c_s$, i.e. where $B_s$ is the magnetic field in the shell and $\mathcal{T} {\rm _s}$ is the contribution of turbulence."1166 Ligure 7 shows how highly sensible to ὃς the fragmentation process ds., Figure 7 shows how highly sensible to $c_s$ the fragmentation process is.1167 Considering LOOSSNeLL and a hot background pressure of 5 νήσο σαι7. the fragmentation takes place even with a low initial transverse velocity. Le. PU)=0.01ΕΙ). i£ c;—1," Considering SNeII and a hot background pressure of 5 $\times$ $^{-10}$ $^{-2}$, the fragmentation takes place even with a low initial transverse velocity, i.e. $v(t_{em})=0.01 V_s(t_{em})$, if $c_s$ $^{-1}$."1168 Increasing the latter by25%. the transverse collapse is very weakened and the fragmentation is prevented.," Increasing the latter by, the transverse collapse is very weakened and the fragmentation is prevented."1169 During the last ten million vears. the evolutions with time of σι and σι are similar. that is. the evolution of &4(/) is mostly driven by the dilution of eu(/) due to the shell expansion.," During the last ten million years, the evolutions with time of $\sigma _0$ and $\tilde \sigma _1$ are similar, that is, the evolution of $\tilde \sigma _1 (t)$ is mostly driven by the dilution of $\sigma _0 (t)$ due to the shell expansion."1170 Equation 35. illustrates the cdillieulty of estimating the sound. speed: of a gas., Equation \ref{eq:cs_mag} illustrates the difficulty of estimating the sound speed of a gas.1171 Ht implies the computations of its cooling history. its magnetic fields and turbulence.," It implies the computations of its cooling history, its magnetic fields and turbulence."1172" Moreover. the high. sensilxlitv of the fragmentation issue to c, (see Fig."," Moreover, the high sensibility of the fragmentation issue to $c_s$ (see Fig."1173 7) shows hat its value must be estimated. with some accuracy., 7) shows that its value must be estimated with some accuracy.1174" Such a task is well bevond the scope of the present work and. in what follows. we adopt c,=Ikm.s5. in agreement witi manv studies of supershell fragmentation (e.g. Comeron Torra 1994. I£hlerosa Palous 2002)."," Such a task is well beyond the scope of the present work and, in what follows, we adopt $c_s=1\,{\rm km.s}^{-1}$, in agreement with many studies of supershell fragmentation (e.g. Comeron Torra 1994, Ehlerova Palous 2002)."1175" The previous section has shown that the shell/swept PGCC may become gravitationally unstable and finally. break into fragments providing that some conditions are fulfilled. c.g. eus)AVsu)& 0.03. οςc9 Lkknestop omagp,c"," The previous section has shown that the shell/swept PGCC may become gravitationally unstable and finally break into fragments providing that some conditions are fulfilled, e.g. $v(t_{em})/V_s(t_{em}) \simeq $ 0.03, $c_s \simeq $ $^{-1}$, $\eta \simeq <\eta _{fg}>$."1176 This is to claim that all supershellswill encounter. such favourable circumstances. but one may expect that at leastsome of them will do.," This is to claim that all supershellswill encounter such favourable circumstances, but one may expect that at least of them will do."1177" ligure 8 presents the results of shell fragmentation simulations for verving values of No P. gg and DSANVGSS). assuming that c,—1kkn.s + and Alba= 0.01 συένο)."," Figure 8 presents the results of shell fragmentation simulations for varying values of $N$, $P_h$, $\eta$ and $v(t_{em})/V_s(t_{em})$, assuming that $c_s$ $^{-1}$ and $\sigma _1 (t_{em}) $ = 0.01 $\sigma _0(t_{em})$."1178" Five hot protogalactic background pressures (2,=—1044.3.2U.101342to4°100 dadvne.cm.7) and 3 SN numbers (V=50. 100. 200) are tested."," Five hot protogalactic background pressures $P_h=10^{-11}, 3.2 \times 10^{-11}, 10^{-10},11793.2 \times 10^{-10}, 10^{-9}$ $^{-2}$ ) and 3 SN numbers $N$ =50, 100, 200) are tested."1180 “Phe upper limit for IN is the maximum number of supernovae that the GC easeous. progenitor can sustain. namely N=2O00 (Le. if NS7200. the absolute value of the cloud. binding energy. is lower than the shell kinetic energv: disruption criterion. Paper 1).," The upper limit for $N$ is the maximum number of supernovae that the GC gaseous progenitor can sustain, namely $N$ =200 (i.e., if $N > 200$, the absolute value of the cloud binding energy is lower than the shell kinetic energy: disruption criterion, Paper I)."1181 Phe lower limit is imposed. by «POCC the sound speed of the οςο material.," The lower limit is imposed by $c^{PGCC}$, the sound speed of the PGCC material."1182 Indeed. the shell is “built” while sweeping the PGCC and such a mass accumulation into a shell requires the velocity of the shell to be larger than the sound. speed of the ambiant medium.," Indeed, the shell is “built” while sweeping the PGCC and such a mass accumulation into a shell requires the velocity of the shell to be larger than the sound speed of the ambiant medium."1183 Therefore. the lower limit to the shell velocity in the PGCC obevs: where Z and peace the temperature (7107 KIS) and the mean molocular weight (2 1.2) of the POGCC. respectively (Pall Rees 1985).," Therefore, the lower limit to the shell velocity in the PGCC obeys: where $T$ and $\mu$ are the temperature $\simeq 10^4$ K) and the mean molocular weight $\simeq 1.2$ ) of the PGCC, respectively (Fall Rees 1985)."1184 Phe number of SNell corresponding to this lower limit of the shell velocity. is Nox50 (leq., The number of SNeII corresponding to this lower limit of the shell velocity is $N \simeq 50$ (Eq.1185 13. Paper D.," 13, Paper I)."1186" Regarcing the upper value of {δι we refer to Alurray Lin (1992) who showed that the hot protogalactic background. pressure depends on the galactocentric distance Das sVhus. D, is ⋠⋅of order ⋅⋅ 7 in. the very inner. Galactic regions (Le. D2 Lkkpe)."," Regarding the upper value of $P_h$, we refer to Murray Lin (1992) who showed that the hot protogalactic background pressure depends on the galactocentric distance $D$ as Thus, $P_h$ is of order $^{-9}$ $^{-2}$ in the very inner Galactic regions (i.e., $D \simeq$ kpc)."1187 1n order to see what is the metallicity achieved in the shells which succeed in forming new stars. iso-metallicity curves corresponding to. L2. 15. 2 and 2.5 (ie. metallicities tvpical of the Galactic halo GC's) are clisplavecl in cach panel of Fig.," In order to see what is the metallicity achieved in the shells which succeed in forming new stars, iso-metallicity curves corresponding to $-$ 1.2, $-$ 1.5, $-$ 2 and $-$ 2.5 (i.e., metallicities typical of the Galactic halo GCs) are displayed in each panel of Fig."1188 8S., 8.1189" For cach couple (N27, ). 54 combinations ΓΕAVuu) and g (ie. 3 initial perturbed velocities 18 numbers of clumps) have been run."," For each couple $N$ $P_h$ ), 54 combinations of $v(t_{em})/V_s(t_{em})$ and $\eta$ (i.e. 3 initial perturbed velocities $\times$ 18 numbers of clumps) have been run."1190 The initial transverse. velocities correspond. to 1. 2 and 3 per cent of the velocity of the shell when it enters the hot background. and the number of clumps ranges from 6 to 40 by step of two.," The initial transverse velocities correspond to 1, 2 and 3 per cent of the velocity of the shell when it enters the hot background and the number of clumps ranges from 6 to 40 by step of two."1191 It appears that the shell is unable to fragment ify=>40., It appears that the shell is unable to fragment if $\eta \ge 40$.1192 When a set of conditions leads to a successful fragmentation. the corresponding point in the CN. D) diagram is marked by a circle as well as by the range of 7 values leading to successful transverse Collapses.," When a set of conditions leads to a successful fragmentation, the corresponding point in the $N$ , $P_h$ ) diagram is marked by a circle as well as by the range of $\eta$ values leading to successful transverse collapses."1193 Depending on whether the shell fragmentation takes place for 0 to 25 per cent. 25 to," Depending on whether the shell fragmentation takes place for 0 to 25 per cent, 25 to"1194later.,later.1195 Since we are more interested in p(o|£). the distribution of c at fixed £. than in the joint distribution pio.£). we weighted cach galaxy by L/o(L£)Vines]. where o(L) is the value of o when that galaxy Iuniisity is muserted iuto the Iuminositv function. aud thens re-]Xotted the joiut distribution of σ aud £.," Since we are more interested in $p(\sigma|L)$, the distribution of $\sigma$ at fixed $L$, than in the joint distribution $p(\sigma,L)$, we weighted each galaxy by $1/[\phi(L){\cal V}_{max}]$, where $\phi(L)$ is the value of $\phi$ when that galaxy's luminosity is inserted into the luminosity function, and then re-plotted the joint distribution of $\sigma$ and $L$."1196 The middle panel of Figure 2 shows the resulting coutour plot., The middle panel of Figure \ref{lvcontours} shows the resulting contour plot.1197 The pawel on the right shows the result of weighting cach galaxy bAN 1[οσα]: i.c.. this panel shows p(L]o).," The panel on the right shows the result of weighting each galaxy by $1/[\phi(\sigma){\cal V}_{max}]$; i.e., this panel shows $p(L|\sigma)$."1198 The dashed ines show from Bernardi et al. (, The dashed lines show from Bernardi et al. (11992003b). where i: is the redshift. (,"2003b), where $z$ is the redshift. ("1200The values 21.15 aud 2.2 are what Bernardi ct al.,The values $-21.15$ and $2.2$ are what Bernardi et al.

Showing the first 1,200 of 10333 lines. Download the file for the rest.