ReadingTimeMachine/rtm-sgt-ocr-v1

1source,target2 We describe how we caleulate matter power spectra for these naiocdels. including the GR. small-scale limit.," We describe how we calculate matter power spectra for these models, including the GR small-scale limit."3 We also describe how we proceed to calculate weak lensing observables from these power spectra., We also describe how we proceed to calculate weak lensing observables from these power spectra.4 In section 77. we present the resulting lensing correlation functions. including realistic errors for future surveys taking into account. shape measurement. noise and cosmic covariance.," In section \ref{results} we present the resulting lensing correlation functions, including realistic errors for future surveys taking into account shape measurement noise and cosmic covariance."5 In section ?? we take the alternative approach of parameterising the non-linear power spectrum. and we investigate how sensitive weak lensing is to these parameters which eo bevond the usual growth parameter.," In section \ref{parameterisation} we take the alternative approach of parameterising the non-linear power spectrum, and we investigate how sensitive weak lensing is to these parameters which go beyond the usual growth parameter."6 We present our conclusions in section ?7.., We present our conclusions in section \ref{conclusions}.7 Throughout this paper we will use a Hat cosmology with the WALAPS|SNeBAO best fit cosmological parameters. which are determined by the background. evolution of the Universe.," Throughout this paper we will use a flat cosmology with the WMAP5+SNe+BAO best fit cosmological parameters, which are determined by the background evolution of the Universe."8" We use a ACDAL backerounel for both ACDAL and ή. in which case we take n4=0.96. =0.71. 04,= ancl ox=O.S1-E0.03 (IXomatsuοἱal.2009).."," We use a $\Lambda$ CDM background for both $\Lambda$ CDM and $f(R)$ in which case we take $n_{\rm s}=0.96$, $h=0.71$, $\Omega _{\rm m}=0.27 \pm 0.02$ and $\sigma_8=0.81 \pm 0.03$ \citep{Komatsu:2008hk}."9" When we use a DOP background. we have 5,=0.998. 5h.=0.66. On,=0.264£0.02 (Fangetal.2008) giving à σε=0.6640.03 for an equivalent ACDAL model,"," When we use a DGP background, we have $n_{\rm s}=0.998$, $h=0.66$, $\Omega _{\rm m}=0.26 \pm 0.02$ \citep{Fang:2008kc} giving a $\sigma_8=0.66 \pm 0.03$ for an equivalent $\Lambda$ CDM model."10 As we have already mentioned. there are two kev phenomena to model in any gravity in order to calculate the matter power spectrum: the expansion history. quantified by the evolution of the Hubble parameter. and the growth history. quantified by the evolution of density perturbations o in the Universe.," As we have already mentioned, there are two key phenomena to model in any gravity in order to calculate the matter power spectrum: the expansion history, quantified by the evolution of the Hubble parameter, and the growth history, quantified by the evolution of density perturbations $\delta$ in the Universe."11 For ACDAL the expansion history is given by the Friedmann equation where //=salla. a is the scale factor and. {1ο is the present day Llubble constant.," For $\Lambda$ CDM, the expansion history is given by the Friedmann equation where $H = \frac{da}{dt} /aH_0$, $a$ is the scale factor and $H_0$ is the present day Hubble constant."12 The growth history is. described. by. the. density perturbation evolution equation together with the Friedmann equation., The growth history is described by the density perturbation evolution equation together with the Friedmann equation.13 At —this point we will limit. ourselves to the regime where density. perturbations evolve linearly., At this point we will limit ourselves to the regime where density perturbations evolve linearly.14 1n this regime we have where primes denote dilferentiation with respect to e., In this regime we have where primes denote differentiation with respect to $a$ .15" This equation is valid for both dark energy. ancl modified eravity models. where Gu is the cllective gravitational constant normalised by the gravitational constant C hence ""-—| for dark energy models. while for mocified. gravity models where 2 is determined by the mocel."," This equation is valid for both dark energy and modified gravity models, where $\tilde{G}_{\rm eff}$ is the effective gravitational constant normalised by the gravitational constant $G$; hence $\tilde{G}_{\rm eff}=1$ for dark energy models, while for modified gravity models where $\beta$ is determined by the model."16 In this paper. we consider DOP (ναctal.2000) and f(A) as examples of modified. gravity models. as the non-linear power spectra have been studied in great detail in these two models using perturbation theory anc N-body simulations.," In this paper, we consider DGP \citep{2000PhLB..485..208D} and $f(R)$ as examples of modified gravity models, as the non-linear power spectra have been studied in great detail in these two models using perturbation theory and N-body simulations."17 For some reviews of modlifieck gravity models see Nojiri&Oclintsov(2006):DurrerMaartens(2008):Ixovama(2008)..," For some reviews of modified gravity models see \cite{Nojiri:2006ri,Durrer:2007re,Koyama:2007rx}."18 In DGP. spacetime has five dimensions. while we live on à 4D brane in the 5D bulk.," In DGP, spacetime has five dimensions, while we live on a 4D brane in the 5D bulk."19 Standard. Model. particles are bound on the 4D brane. asis gravity on small scales: however on large scales gravity leaks olf the brane causing late time acceleration.," Standard Model particles are bound on the 4D brane, asis gravity on small scales; however on large scales gravity leaks off the brane causing late time acceleration."20 The scale of the transition [rom 4D to 5D eravity is governed. by the crossover scale. re=(1Ou) I.," The scale of the transition from 4D to 5D gravity is governed by the crossover scale, $r_{\rm c}=(1-\Omega _{\rm m})^{-1}$ ."21 The extra dimension contributes a further term to the Friedmann equation whose amplitude is governed by re: The growth history is also altered. giving (Ixovama&Martens2006) In f(H) gravity models the Einstein-Llilbert action is moclified to include an arbitrary function of the Ricci scalar. HI.," The extra dimension contributes a further term to the Friedmann equation whose amplitude is governed by $r_{\rm c}$ : The growth history is also altered, giving \citep{Koyama:2005kd}22 In $f(R)$ gravity models the Einstein-Hilbert action is modified to include an arbitrary function of the Ricci scalar, $R$."23" dn this studywe use an fC?) function of the form Sawicki2007b) where 2 is the Ricci scalar. Ze) is the present day Iicci scalar and fg,=aId "," In this studywe use an $f(R)$ function of the form \citep{Hu:2007nk}24 where $R$ is the Ricci scalar, $R_0$ is the present day Ricci scalar and $f_{\rm R_0}=\left.\frac{df}{dR}\right|_{ R=R_0}$."25"We use [/5,|=10 7. which has been found to fit with cluster constraints (Schmidtetal... 2009). to give a background. evolution. which is approximately ACDAL to sub-pereent level."," We use $\left|f_{\rm R_0}\right| = 10^{-4}$ , which has been found to fit with cluster constraints \citep{Schmidt:2009am}, to give a background evolution which is approximately $\Lambda$ CDM to sub-percent level."26 This allows us to use the ACDAL Friedmann equation and only alter the density evolution equation (Lueetal.2004b:Zhang2006:lxovamaetal.2009) with where A is the dimensionless wavenumber defined: as ΑΟHo). kis the wavenumber and e is the speed of light.," This allows us to use the $\Lambda$ CDM Friedmann equation and only alter the density evolution equation \citep{Lue:2004rj, Zhang:2005vt, Koyama:2009me} with where $\bar{k}$ is the dimensionless wavenumber defined as $k(c/H_0)$, $k$ is the wavenumber and $c$ is the speed of light."27 For modified gravity to agree with solar. svsteni observations it must approach a GRO solution on small scales., For modified gravity to agree with solar system observations it must approach a GR solution on small scales.28 This means that the non-linear power spectrum must be an interpolation of the mocdified gravity non-linear power spectrum with nomechanism to obtain the Gh result onsmall scales. P4544.oud. 2). and the GR non-linearpower spectrum with the same expansion history as the moclificc gravity model.Lop(hk. z).," This means that the non-linear power spectrum must be an interpolation of the modified gravity non-linear power spectrum with nomechanism to obtain the GR result onsmall scales, $P_{\rm non-GR}(k,z)$ , and the GR non-linearpower spectrum with the same expansion history as the modified gravity model,$P_{\rm GR}(k,z)$ ."29 A fitting formula for this interpolation was proposed by Llu&Sawicki (2007a)::, A fitting formula for this interpolation was proposed by \cite{Hu:2007pj}: :30from the emission at 8 and 24 μπι if we find a good correlation between these and Εν.,"from the emission at 8 and $24\,\mu$ m if we find a good correlation between these and $_{TIR}$."31 For the sample of resolved sources at 70 and 160 jm we derive Εμ and its relationship to the emission at 8 and 24 jm using the IRAC 8.0 gam photometry.," For the sample of resolved sources at 70 and 160 $\mu$ m we derive $_{TIR}$ and its relationship to the emission at 8 and $24\,\mu$ m using the IRAC 8.0 $\mu$ m photometry."32 This relation has been inferred previously but it might depend on the properties of the galaxy considered and on the scale examined (e.g.??)..," This relation has been inferred previously but it might depend on the properties of the galaxy considered and on the scale examined \citep[e.g.][]{2005ApJ...628L..29E,2006ApJ...648..987P}."33" To find a relation between the 8and 24 jum emission and Ly;g. ? plotted the {νε{ον ratio as a function of the 8/24 uim flux ratio for their sample of stellar complexes in the inner disk and derived a linear fit: where Lo is vF,x4xD7."," To find a relation between the 8and $24\,\mu$ m emission and $L_{TIR}$, \citet{2009A&A...493..453V} plotted the $L_{24}/L_{TIR}$ ratio as a function of the 8/24 $\mu$ m flux ratio for their sample of stellar complexes in the inner disk and derived a linear fit: where $L_{24}$ is $\nu F_\nu \times 4 \pi D^2 $."34 We have checked that the above relation also holds in the outer disk re. that the numerical coefficients of the above relation for sources in the outer disk are. within the errors. compatible with those relative to sources in the inner disk.," We have checked that the above relation also holds in the outer disk i.e. that the numerical coefficients of the above relation for sources in the outer disk are, within the errors, compatible with those relative to sources in the inner disk."35 Thus. in this study we adopt the relation of ? to compute Ly; from the 8 and 24 jm fluxes.," Thus, in this study we adopt the relation of \citet{2009A&A...493..453V} to compute $L_{TIR}$ from the 8 and $24\,\mu$ m fluxes."36 An error of 0.10 mag is added in quadrature to the photometric error to compute the uncertainties in the 8 yam flux., An error of 0.10 mag is added in quadrature to the photometric error to compute the uncertainties in the 8 $\mu$ m flux.37 The IRAC 84m map ts smaller than the 24 jm mosaic (see Fig.," The IRAC $8\,\mu$ m map is smaller than the $24\,\mu$ m mosaic (see Fig."38 2 upper left panel) and 57 of the 915 sources fall outside it., \ref{map} upper left panel) and 57 of the 915 sources fall outside it.39 For these sources and for the few (25) without a reliable 8.0 uim photometry (error<0.5 mag) we use the relation to derive the flux at Sym which is needed to caleulate Ly jp., For these sources and for the few (25) without a reliable 8.0 $\mu$ m photometry $<$ 0.5 mag) we use the relation to derive the flux at $\mu$ m which is needed to calculate $L_{TIR}$ .40 These coefficients are derived from a least square fit to the sources detected in both bands., These coefficients are derived from a least square fit to the sources detected in both bands.41universal. but depends ou both ealaxy type aud the local environment.,"universal, but depends on both galaxy type and the local environment."42 Both Chniistlei (7)) and Zablucloff A[Dulchaey (?)) &ud that the change iu the optical LF shape with environment is due alinost exclusively to the NEL ealaxies., Both Christlein \shortcite{Christ}) ) and Zabludoff Mulchaey \shortcite{ZM00}) ) find that the change in the optical LF shape with environment is due almost exclusively to the NEL galaxies.43 We confi that the difference in the infrared cluster LF. rolative to the field. is donünated bv this population.," We confirm that the difference in the infrared cluster LF, relative to the field, is dominated by this population."44 It is interesting that the faint eud slope of the NEL galaxies iu clusters. as parametrized bw a. is similar to that of the overall field faint end slope.," It is interesting that the faint end slope of the NEL galaxies in clusters, as parametrized by $\alpha$, is similar to that of the overall field faint end slope."45 One oeiterpretationu for this smüluitv is that the bulk of the cluster population is built up bv accretiug field. &alaxies with little effect other than the cessation of star formation. as modeled for example by Balogh. Navarro Moris (?]).," One interpretation for this similarity is that the bulk of the cluster population is built up by accreting field galaxies with little effect other than the cessation of star formation, as modeled for example by Balogh, Navarro Morris \shortcite{infall}) )."46 ILowever. from ouly the agreement of the LEs. we caunot exclude scenarios in which the similar initial progeutors of faint galaxies imn various environments beeau forming stars at different times (the cluster ones earlier by virtue of originating near a large mass perturbation).," However, from only the agreement of the LFs, we cannot exclude scenarios in which the similar initial progentors of faint galaxies in various environments began forming stars at different times (the cluster ones earlier by virtue of originating near a large mass perturbation)."47 In the latter scenario. one expects the field LF to evolve to the cluster LF iu time. as gas is coustuned by ealaxies and their star formation stops.," In the latter scenario, one expects the field LF to evolve to the cluster LF in time, as gas is consumed by galaxies and their star formation stops."48 Iu Figure LL we show the fraction of NEL galaxies as a function of luminosity for the global sample., In Figure \ref{fig-fnel} we show the fraction of NEL galaxies as a function of luminosity for the global sample.49 This Figure shows a strong treud that. wafortunately. is ditiicult to interpret because metallicity aud extinction also correlate with huuiuositv in a way that makes |OII| stronger in low Inmuinosity galaxies (0.8.. Jansen ?)).," This Figure shows a strong trend that, unfortunately, is difficult to interpret because metallicity and extinction also correlate with luminosity in a way that makes [OII] stronger in low luminosity galaxies (e.g., Jansen \shortcite{Jansen01}) )."50 We cannot infer the extent to which this relation iudicates an iuhereut correlation between ealaxy mass aud instautancous star formation rate — a correlation that would be an iniportaut test of galaxy formation models., We cannot infer the extent to which this relation indicates an inherent correlation between galaxy mass and instantaneous star formation rate — a correlation that would be an important test of galaxy formation models.51 In particular. a variation in the fraction of currently star forming galaxies with mass could indicate that the dominant mode of star formation (.c.. burst-like or coutinuous) for a galaxy in the field is mass dependent (c.e.. INauffinann. Charlot," In particular, a variation in the fraction of currently star forming galaxies with mass could indicate that the dominant mode of star formation (i.e., burst-like or continuous) for a galaxy in the field is mass dependent (e.g., Kauffmann, Charlot"52Compact groups (hereafter CGs) of galaxies are. small associations of galaxies on the sky characterized by a few members. of the order of four to eight. by a relatively small velocity dispersion. of the order of200kms!.. that are separated on the sky by an average distance comparable to the diameter of individual galaxies.,"Compact groups (hereafter CGs) of galaxies are small associations of galaxies on the sky characterized by a few members, of the order of four to eight, by a relatively small velocity dispersion, of the order of, that are separated on the sky by an average distance comparable to the diameter of individual galaxies."53 Under the assumption that CGs are gravitationally bound objects free from other external influences. one expects these CGs. owing to their mutual interactions. to evolve rapidly. following several violent interactions among the member galaxies. and to form a single isolated early type galaxy in a short time interval compared with the Hubble time (see for example Barnes. Nevertheless. despite many multiwavelength observations and intensive analyses have confirmed that many compact groups are gravitationally bound objects (e.g. Verdes-Montenegro et al.," Under the assumption that CGs are gravitationally bound objects free from other external influences, one expects these CGs, owing to their mutual interactions, to evolve rapidly, following several violent interactions among the member galaxies, and to form a single isolated early type galaxy in a short time interval compared with the Hubble time (see for example Barnes, Nevertheless, despite many multiwavelength observations and intensive analyses have confirmed that many compact groups are gravitationally bound objects (e.g. Verdes-Montenegro et al."54 2001 (VMOI). Ponman et al..," 2001 (VM01), Ponman et al.,"55 1996: Mendes de Oliveira et al..," 1996; Mendes de Oliveira et al.,"56. 1994 (MDO): Hickson et al..," 1994 (MDO); Hickson et al.,"57 1992). the picture that has emerged is remarkably more complex than the one suggested by earlier studies.," 1992), the picture that has emerged is remarkably more complex than the one suggested by earlier studies."58 To begin with. members of isolated compact groups had only a small fraction of strongly interacting galaxies. of the order of 7% (MDO. 94). in contrast to the expectations of N-body simulations.," To begin with, members of isolated compact groups had only a small fraction of strongly interacting galaxies, of the order of $\%$ (MDO, 94), in contrast to the expectations of N-body simulations."59 However. evidence of gentler interactions (e.g. gas stripping) was detected in almost half of the member galaxies. implying that there was some kind of influence of the group environment on the evolution of its members.," However, evidence of gentler interactions (e.g. gas stripping) was detected in almost half of the member galaxies, implying that there was some kind of influence of the group environment on the evolution of its members."60 Other correlations. e.g. the one between velocity dispersion and dominant morphological type in groups. and that between crossing time and spiral fraction (Hickson et al..," Other correlations, e.g. the one between velocity dispersion and dominant morphological type in groups, and that between crossing time and spiral fraction (Hickson et al.,"61 1992) implied that CGs follow an evolutionary path that leads to several possible endings., 1992) implied that CGs follow an evolutionary path that leads to several possible endings.62 Compact groups were then proposed to merge and to form an isolated early type galaxy. or. depending on the original mass. a fossil group.," Compact groups were then proposed to merge and to form an isolated early type galaxy, or, depending on the original mass, a fossil group."63 Alternatively. it was proposed that their lifetimes were much longer than predicted by early numerical simulations owing to a massive halo of dark matter stabilizing the compact group for a long time.," Alternatively, it was proposed that their lifetimes were much longer than predicted by early numerical simulations owing to a massive halo of dark matter stabilizing the compact group for a long time."64 This excluded a short lifetime and explained the difficulty in identifying the final merging product of a Not all compact groups were. however. found to be às isolated on the sky as originally supposed (see for example de Carvalho et al..," This excluded a short lifetime and explained the difficulty in identifying the final merging product of a Not all compact groups were, however, found to be as isolated on the sky as originally supposed (see for example de Carvalho et al.,"65 1994; Ribeiro et al..," 1994; Ribeiro et al.,"66 1998)., 1998).67 Some of them were found to be quite close to clusters of galaxies whose richness seems to vary with redshift (Andernach Coziol. 2005). while the remainder could be divided into three categories.," Some of them were found to be quite close to clusters of galaxies whose richness seems to vary with redshift (Andernach Coziol, 2005), while the remainder could be divided into three categories."68 These were named: (1)groups. 1.9. a larger than. previously expected galaxy concentration: (2) a configuration. i.e. a central concentration within a looser distribution of galaxies; and finally. (3) compact groups that complied with the original definitions. ie were truly isolated and gravitationally bound dense structures.," These were named: (1), i.e. a larger than previously expected galaxy concentration; (2) a configuration, i.e. a central concentration within a looser distribution of galaxies; and finally, (3) compact groups that complied with the original definitions, i.e were truly isolated and gravitationally bound dense structures."69" This led many authors to propose that CGs are a local universe phenomenon in a biased cold-dark-matter galaxy formation model (West. 1989, Andernach Coziol 2005. to cite a few)."," This led many authors to propose that CGs are a local universe phenomenon in a biased cold-dark-matter galaxy formation model (West, 1989, Andernach Coziol 2005, to cite a few)."70 Larger scale structures would then form first. leaving smaller associations such as compact groups to form last. with or shortly before field galaxies.," Larger scale structures would then form first, leaving smaller associations such as compact groups to form last, with or shortly before field galaxies."71 The different kind of groups observed were just different structures at different spatial scales and formation In constrast. Einasto et al..," The different kind of groups observed were just different structures at different spatial scales and formation In constrast, Einasto et al.,"72 2003. showed that loose groups of galaxies close to large-scale structures are on average more massive and have a larger velocity dispersion than those that are more isolated on the sky.," 2003, showed that loose groups of galaxies close to large-scale structures are on average more massive and have a larger velocity dispersion than those that are more isolated on the sky."73 According to these authors. this is evidence that the large-scale gravitational field responsible for the formation of rich clusters enhances the evolution of neighbouring poor systems.," According to these authors, this is evidence that the large-scale gravitational field responsible for the formation of rich clusters enhances the evolution of neighbouring poor systems."74 A larger velocity dispersion implies a higher mass. 1.8. that an environmental enhancement of mass is observed.," A larger velocity dispersion implies a higher mass, i.e. that an environmental enhancement of mass is observed."75 This in turn is interpreted as direct evidence of the hierarchical formation of galaxies and clusters in à network of filaments connecting high density knots of the cosmic mass To shed some light on the evolutionary path of CGs and on their relation to environment. and in order to understand what is the role of CGs in the evolution of their member galaxies and the larger-scale structures. 1t 1s Imperative to extend to higher," This in turn is interpreted as direct evidence of the hierarchical formation of galaxies and clusters in a network of filaments connecting high density knots of the cosmic mass To shed some light on the evolutionary path of CGs and on their relation to environment, and in order to understand what is the role of CGs in the evolution of their member galaxies and the larger-scale structures, it is imperative to extend to higher"76 (e.g2): (??)..," \citep[e.g][]{1998ApJ...501L..89B}; \citep{1998Natur.394..344W,2002IAUC.7867....1M}. \citep{1982Natur.300..615B}."77 (?).. ? ? r—1nodes) ? (see?.foradditionalcliscussion)..," \citet{1999ApJ...516..307A} \citet{1998ApJ...501L..89B} $r-$ \citet{1999ApJ...517..328L} \citep[see][for additional78discussion]{2000ApJ...534L..75A}."79 r—qmode. ?.—r—imocles ?.. T ?7 7.?. ?..," $r-$ \citet{1999ApJ...517..328L} $r-$ \citet{2001astro.ph.10487S}, \citet{2001ApJ...549.1011W}80 \citet{Arra02} \citet{1999ApJ...516..307A} \citet{1998ApJ...501L..89B} \citet{1999ApJ...517..328L}."81Figureby 3 compares equationre« (5)) with the exact equations from Agol (2002).,Figure 3 compares equation\ref{edge}) ) with the exact equations from Agol (2002).82 For smaller source size. the edge has a Napoleonic bicorne shape. while larger sources look more like a tophat.," For smaller source size, the edge has a Napoleonic bicorne shape, while larger sources look more like a tophat."83 The straight-edge approximation improves as the source becomes larger. and approaches a step function as R./Rp»x.," The straight-edge approximation improves as the source becomes larger, and approaches a step function as $R_*/R_E 84\rightarrow \infty$."85" When the lens has afinite size. rj2Rr/Rg>0. the obscuration of the images by the lensgives where ¢2 r.—z. s;= sign(G). s,=signi"," When the lens has afinite size, $r_L=R_L/R_E > 0$, the obscuration of the images by the lensgives where $\zeta=r_*-z$ , $s_\zeta = sign(\zeta)$ , $s_r=sign(r_L-1)$ ."86" (Hubble1926) Figurel)). thin disk of stars that show a range of spiral patterns or ""arms""."," \citep{Hubble1926} \ref{tuningfork}) thin disk of stars that show a range of spiral patterns or “arms""."87 Hubble ordered disk galaxies based on the tightness of these spiral arms and the size of the central bulge., Hubble ordered disk galaxies based on the tightness of these spiral arms and the size of the central bulge.88 He had two distinct populations of disk galaxies. namely with and without a central bar-like (or linear) structure.," He had two distinct populations of disk galaxies, namely with and without a central bar-like (or linear) structure."89" At the point where these different classifications met (for spirals with the largest bulges. and tightest wound arms). Hubble placed ""lenticular"" galaxies which at the time were hypothetical disk galaxies with very large bulges and no spiral arms — they have since been found."," At the point where these different classifications met (for spirals with the largest bulges, and tightest wound arms), Hubble placed “lenticular"" galaxies which at the time were hypothetical disk galaxies with very large bulges and no spiral arms – they have since been found."90 It is a common misconception that Hubble believed the “tuning fork” diagram was an evolutionary sequence. with elliptical galaxies on the left evolving along the sequence to form disk galaxies.," It is a common misconception that Hubble believed the “tuning fork"" diagram was an evolutionary sequence, with elliptical galaxies on the left evolving along the sequence to form disk galaxies."91" In fact Hubble advised that “temporal connotations are made at one's peril” in an early defence of the classification sequence (Hubble1927).. going on to say that he set up the classification ""without prejudice to theories of [galaxy] evolution""."," In fact Hubble advised that “temporal connotations are made at one's peril"" in an early defence of the classification sequence \citep{Hubble1927}, going on to say that he set up the classification “without prejudice to theories of [galaxy] evolution""."92" This misconception about Hubble's beliefs probably arose due to his suggestion of the use of ""early"" and “late” types to describe the progression towards the right along the sequence (although he discussed that this nomenclature was simply for convenience and borrowed terminology commonly used for stellar classification. (Hubble 1926)))."," This misconception about Hubble's beliefs probably arose due to his suggestion of the use of “early"" and “late"" types to describe the progression towards the right along the sequence (although he discussed that this nomenclature was simply for convenience and borrowed terminology commonly used for stellar classification, \citep{Hubble1926}) )."93" Astronomers still call elliptical galaxies ""early"" types and disk galaxies ""late"" type galaxies. although we now know that most ""late"" type galaxies have much younger stellar populations (ironically more “early type"" stars) than most ""early"" type galaxies."," Astronomers still call elliptical galaxies “early"" types and disk galaxies “late"" type galaxies, although we now know that most “late"" type galaxies have much younger stellar populations (ironically more “early type"" stars) than most “early"" type galaxies."94 Since Hubble. there have been several updates to his classification scheme (forarecentreviewseeButa2011) but key features have remained unchanged.," Since Hubble, there have been several updates to his classification scheme \citep[for a recent review see][]{Buta2011} but key features have remained unchanged."95 What has changed dramatically i$ the number of galaxies catalogued and requiring classification., What has changed dramatically is the number of galaxies catalogued and requiring classification.96 Before the advent of digital detectors in astronomy. astronomers could just visually classify the galaxies they saw via their telescopes and/or on photographie plates.," Before the advent of digital detectors in astronomy, astronomers could just visually classify the galaxies they saw via their telescopes and/or on photographic plates."97 New astronomers were trained to follow the classification rules and provided detailed morphologies for thousands of galaxies., New astronomers were trained to follow the classification rules and provided detailed morphologies for thousands of galaxies.98 Several large catalogues of nearby galaxies with such classifications exists (e.g. The Hubble Atlas of Galaxies (Sandage1961).. or the Third Reference Catalogue of Bright Galaxies (RC3). (deVaucouleursetal. 1991))). and many of these classifications are collected in the," Several large catalogues of nearby galaxies with such classifications exists (e.g. The Hubble Atlas of Galaxies \citep{Sandage1961}, or the Third Reference Catalogue of Bright Galaxies (RC3), \citep{deVaucouleurs1991}) ), and many of these classifications are collected in the"99and circular polarizations.,and circular polarizations.100 We construct such histograms for all our examples. but here we show only the Poincaré sphere [or most of them.," We construct such histograms for all our examples, but here we show only the Poincaré sphere for most of them."101 The simulation shown in Figure 2. reprocluces the example shown in the lower panel in Figure 1 reasonably well., The simulation shown in Figure \ref{fig1} reproduces the example shown in the lower panel in Figure \ref{figES} reasonably well.102 Lt is clear that this ancl similar looking examples may be simulated in terms of two orthogonal modes with random. spreads about the two polarizations., It is clear that this and similar looking examples may be simulated in terms of two orthogonal modes with random spreads about the two polarizations.103 However. it is also clear that additional assumptions need to be mace to simulate the example shown in the upper panel in Figure 1..," However, it is also clear that additional assumptions need to be made to simulate the example shown in the upper panel in Figure \ref{figES}."104 We attempt to do so by relaxing one-hy-one some of the assumptions made in the simulation shown in Figure :, We attempt to do so by relaxing one-by-one some of the assumptions made in the simulation shown in Figure \ref{fig1}.105 Also note that despite the variances in the polarizations being the same. and the mean intensities being the same. for both mocles. the distributions on the Poincaré sphere are dilferent: this difference is particularly notable in the histogram for PA.," Also note that despite the variances in the polarizations being the same, and the mean intensities being the same, for both modes, the distributions on the Poincaré sphere are different; this difference is particularly notable in the histogram for PA."106 ‘These clilferences are due entirely to the cdillercnt statistics for the intensities of the two modes: in this case the highest intensities are dominated by mode 2. with £afy} and 15)By=qp1.256879n[2 ," These differences are due entirely to the different statistics for the intensities of the two modes: in this case the highest intensities are dominated by mode 2, with $\langle I_2\rangle=\langle I_1\rangle$ and $\langle I_2^2\rangle=1.25\langle I_1^2\rangle$ ."107The example illustrated in Figure 8 differs from that in Figure 2 in that the variances in the polarizations of the two mocdes are dillerent. and the mean of the ratio of the intensities of the two modes is increased slightly ereater than unity (hy 1.1).," The example illustrated in Figure \ref{fig2} differs from that in Figure \ref{fig1} in that the variances in the polarizations of the two modes are different, and the mean of the ratio of the intensities of the two modes is increased to slightly greater than unity $\Iot=1.1$ )."108" For mode 1. the choice a,0.05. b,0.15. implies that the spread. is larger on the Poincaré sphere in the vertical than the horizontal direction: the surfaces of constant probability are ellipses with axial ratio byfa,= 3. favoring circular over linear »olarization."," For mode 1, the choice $a_1=0.05$, $b_1=0.15$, implies that the spread is larger on the Poincaré sphere in the vertical than the horizontal direction; the surfaces of constant probability are ellipses with axial ratio $b_1/a_1=3$ , favoring circular over linear polarization."109 For mode 2. the choice a»=002. bs=0.01. corresponds to a smaller spread than for mode 1. and such hat the probability ellipses have axial ratio bsfz=1/2. avoring linear over circular polarization.," For mode 2, the choice $a_2=0.02$, $b_2=0.01$, corresponds to a smaller spread than for mode 1, and such that the probability ellipses have axial ratio $b_2/a_2=1/2$, favoring linear over circular polarization."110 The cllect. shown in Figure 3.. is to spread out the points around the mean for mode 1 (on the right) much more strongly than for mode 2 (on the left).," The effect, shown in Figure \ref{fig2}, is to spread out the points around the mean for mode 1 (on the right) much more strongly than for mode 2 (on the left)."111 lt is clear that while such spreacling might be one ingredient in attempting to simulate the upper panel in Figure 1.. it cannot explain the concentration of the points around a broad annulus. rather than a central modal point.," It is clear that while such spreading might be one ingredient in attempting to simulate the upper panel in Figure \ref{figES}, it cannot explain the concentration of the points around a broad annulus, rather than a central modal point."112 The increase in Ay from unity in Figure 2 to hy l.lin Figure 3 is relatively unimportant: this ratio becomes more important in Figures 4 to 7 where we attempt to simulate the annulus in Figure L.., The increase in $\Iot$ from unity in Figure \ref{fig1} to $\Iot=1.1$ in Figure \ref{fig2} is relatively unimportant; this ratio becomes more important in Figures \ref{fig3} to \ref{fig6} where we attempt to simulate the annulus in Figure \ref{figES}.113 The example illustrated in Figure 4. differs from that in Figure 3. in that the spread in the ratio of intensities is much smaller: the parameter A.0.5 in Ligure 3 is replaced by As=0.05 in Figure. 4.., The example illustrated in Figure \ref{fig3} differs from that in Figure \ref{fig2} in that the spread in the ratio of intensities is much smaller: the parameter $\dfIt=0.5$ in Figure \ref{fig2} is replaced by $\dfIt=0.05$ in Figure \ref{fig3}.114 This causes the istribution of points to become even more strongly spread out. and for the concentration of points around the means o disappear.," This causes the distribution of points to become even more strongly spread out, and for the concentration of points around the means to disappear."115 In interpreting this. first consider a case (not ga10wn) where there is no spread in the ratio of the intensities X mode 2 to that in. mocde Ay=1. do," In interpreting this, first consider a case (not shown) where there is no spread in the ratio of the intensities of mode 2 to that in mode 1, $\Iot=1$, $\dfIt\to0$."116 ‘Then. on summing over the Stokes parameters for the two moces 10 mean polarizations cancel.," Then, on summing over the Stokes parameters for the two modes the mean polarizations cancel."117 The cancellation is not exact because the parameters for cachmode correspond. to two ilferent. choices of random. numbers., The cancellation is not exact because the parameters for eachmode correspond to two different choices of random numbers.118 The net. polarization is determined. by the difference between the two moces. and the degree of polarization is necessarily small. (Qi|QS|i0C371.AELSY]mEP 91. dueto distribution(Qe.(Q»το)zzΦινινV4.," The net polarization is determined by the difference between the two modes, and the degree of polarization is necessarily small, $[(Q_1+Q_2)^2+(U_1+U_2)^2+(V_1+V_2)^2]^{1/2}\ll I_1+I_2=2I_1$ , due to $(Q_2,U_2,V_2)\approx-Q_1,U_1,V_1$."119 This leads broad of polarization on the Poincaré. sphere., This leads to a broad distribution of polarization on the Poincaré sphere.120 The difference between Figure 8.. Ae=0.5 and Figure 4.. with iab the latter is elfectively indistinguishable from the case As» Oin which the mean polarizations cancel exactly.," The difference between Figure \ref{fig2}, $\dfIt=0.5$ and Figure \ref{fig3}, , with $\dfIt=0.05$ is that the latter is effectively indistinguishable from the case $\dfIt\to0$ in which the mean polarizations cancel exactly."121 This example adds a further ingredient that is plausibly needed in the interpretation of the upper panel in Figure L:: spreading out of the points due to near equality of the intensities in the two mocdoes., This example adds a further ingredient that is plausibly needed in the interpretation of the upper panel in Figure \ref{figES}: spreading out of the points due to near equality of the intensities in the two modes.122 However. the associated. loss of concentration around the mean polarizations for the two modes is not consistent with the observations. and a further assumption is needed to overcome this.," However, the associated loss of concentration around the mean polarizations for the two modes is not consistent with the observations, and a further assumption is needed to overcome this."123 Aefore considering how this can be achieved. we relax the assumption of orthogonality.," Before considering how this can be achieved, we relax the assumption of orthogonality."124 The example shown in Figure 5. dilfers from Figure 4. only in that the centroid for mode 2 is not orthogonal to that for mode 1I. and is displaced from the antipodal point by λε= 37. ΔΣ=10," The example shown in Figure \ref{fig4} differs from Figure \ref{fig3} only in that the centroid for mode 2 is not orthogonal to that for mode 1, and is displaced from the antipodal point by $\dPAo=3^\circ$ , $\dEllAo=10^\circ$."125 Vhis introduces a favored. direction on the Poincaré sphere. and. the polarization points tend to form an annulus around this preferred. direction.," This introduces a favored direction on the Poincaré sphere, and the polarization points tend to form an annulus around this preferred direction."126 This annulus »ovides a natural explanation for the spreading apparent in he observational examplein the upper panel of Figure 1.., This annulus provides a natural explanation for the spreading apparent in the observational examplein the upper panel of Figure \ref{figES}. .127 Llowever. the absence of a noticeable concentration of pointsaround the mean polarization for mode 2 is inconsistent with he observations.," However, the absence of a noticeable concentration of pointsaround the mean polarization for mode 2 is inconsistent with the observations."128 Note that theannulus appears when the ratio of the, Note that theannulus appears when the ratio of the129"to establish constraints on GRB rates and beaming factors. ?,,","to establish constraints on GRB rates and beaming factors. \citet{Soderberg2006},"130" for example, conclude from late time radio observations of 68 local type Ibc supernovae (SNe) that less than -10% of such SNe are associated with GRBs, and constrain the GRB beaming factor to be ((1—c0s0;)!)S10*."," for example, conclude from late time radio observations of 68 local type Ibc supernovae (SNe) that less than $\backsim 10 \%$ of such SNe are associated with GRBs, and constrain the GRB beaming factor to be $\left< (1 -131 \cos{\theta_j})^{-1} \right> \lesssim 10^4$."132" A lower limit to the beaming factor of (a—cos8,1)=,13 is provided by ?.."," A lower limit to the beaming factor of $\left< (1 - \cos{\theta_j})^{-1} \right> \gtrsim13313$ is provided by \citet{Levinson_etal_2002_ApJ}."134" Such estimates require a model describing the shape of off-axis light-curves, and are therefore sensitive to model assumptions."," Such estimates require a model describing the shape of off-axis light-curves, and are therefore sensitive to model assumptions."135" Eventually, comparing observations and detailed simulations like the one described in this paper will place the most accurate observational limits on orphan afterglow characteristics."," Eventually, comparing observations and detailed simulations like the one described in this paper will place the most accurate observational limits on orphan afterglow characteristics."136 A large number of simulations is required to fully explore the afterglow parameter space., A large number of simulations is required to fully explore the afterglow parameter space.137" We can however, use the single simulation of this paper that has typical values for the explosion parameters to confirm the result from ? that their sample of 68 supernovae observations are all significantly fainter than a standard afterglow viewed off-axis."," We can however, use the single simulation of this paper that has typical values for the explosion parameters to confirm the result from \citet{Soderberg2006} that their sample of 68 supernovae observations are all significantly fainter than a standard afterglow viewed off-axis."138" This confirmation is shown in fig. 11,,"," This confirmation is shown in fig. \ref{supernovae_figure},"139" where we have plotted 66 supernovae radio upper limits (omitting SN 1984L and SN1954A, which were not observed at 8.46 GHz, from the 68) together with our off-axis simulated light curves."," where we have plotted 66 supernovae radio upper limits (omitting SN 1984L and SN1954A, which were not observed at 8.46 GHz, from the original 68) together with our off-axis simulated light curves."140" Note originalthat the jet half opening angle in our simulation is 11.5?, whereas ? use 5?."," Note that the jet half opening angle in our simulation is $11.5^\circ$, whereas \citet{Soderberg2006} use $5^\circ$."141" The fact that the early time flux received by an off-axis observer is actually stronger than (as shown in section 3.2,, where model and analyticallysimulation are expectedcompared directly) only strengthens the case made by Soderberg et al."," The fact that the early time flux received by an off-axis observer is actually stronger than analytically expected (as shown in section \ref{off_axis_comparison_section}, where model and simulation are compared directly) only strengthens the case made by Soderberg et al."142 A possible caveat to the above is that our simulation light curves do not include the effect of synchrotron self-absorption., A possible caveat to the above is that our simulation light curves do not include the effect of synchrotron self-absorption.143" Although we cannot completely rule out that this plays a role without actually calculating it, we can nevertheless look at the effect of self-absorption on the model light curves, having already established thatmodel and simulation lead to at least qualitatively similar light curves in section 3.2.."," Although we cannot completely rule out that this plays a role without actually calculating it, we can nevertheless look at the effect of self-absorption on the model light curves, having already established thatmodel and simulation lead to at least qualitatively similar light curves in section \ref{off_axis_comparison_section}."144 In Fig., In Fig.145" 12 we show model light curves with and without self-absorption, calculated as explained in the appendix."," \ref{selfabsorption_figure} we show model light curves with and without synchrotron self-absorption, calculated as explained in the appendix."146 synchrotronThe figure shows that the effect of self-absorption is initially significant for an on-axis observer but becomes less pronounced for observers further off-axis., The figure shows that the effect of self-absorption is initially significant for an on-axis observer but becomes less pronounced for observers further off-axis.147" For an observer at 90°, the light curves with and without self-absorption are effectively identical."," For an observer at $90^\circ$, the light curves with and without self-absorption are effectively identical."148" Aside from the minimal differences due to the analytical model assumptions, the main difference between this figure and fig."," Aside from the minimal differences due to the analytical model assumptions, the main difference between this figure and fig."149 1 from et al., 1 from Soderberg et al.150 is due to the different jet opening angles., is due to the different jet opening angles.151" ForSoderberg our wider jet opening angle, only the two earliest supernovae lie clearly above the 90? curve."," For our wider jet opening angle, only the two earliest supernovae lie clearly above the $90^\circ$ curve."152" In this paper we present broadband GRB afterglow light curves calculated assuming synchrotron emission from a high-resolution relativistic jet simulation in 2D. We have expanded the work presented in ?— to include observers positioned off the symmetry axis, both at small and large angles."," In this paper we present broadband GRB afterglow light curves calculated assuming synchrotron emission from a high-resolution relativistic jet simulation in 2D. We have expanded the work presented in \citet{Zhang2009} to include observers positioned off the jet symmetry axis, both at small and large angles."153" For the jet simulationjet we have used the adaptive-mesh-refinement code, starting from the Blandford-McKee analytical solution and letting the jet evolve until it has reached the Sedov-Taylor stage and has decollimated into a nearly spherical outflow."," For the jet simulation we have used the adaptive-mesh-refinement code, starting from the Blandford-McKee analytical solution and letting the jet evolve until it has reached the Sedov-Taylor stage and has decollimated into a nearly spherical outflow."154 We have implemented synchrotron radiation as described in ?.., We have implemented synchrotron radiation as described in \citet{Sari1998}. .155" When put in the context of analytical light curve estimates, our simulations show the following:"," When put in the context of analytical light curve estimates, our simulations show the following:"156describe the microlensing effect. (5.H4) for a given SBP.,"describe the microlensing effect $b,R_0$ ) for a given SBP."157 Considering all the data for GRD 000301C' in Figure 1. there are 11 parameters without microlensing. aud 13 free parameters with microlensing using theoreticallv-caleulated SBPs.," Considering all the data for GRB 000301C in Figure 1, there are $11$ parameters without microlensing, and $13$ free parameters with microlensing using theoretically-calculated SBPs."158 For comparison purposes. we also fit the data to the model adopted by GLS: we assume that the source emission is confined (o a thin ring of fractional width VW.," For comparison purposes, we also fit the data to the model adopted by GLS: we assume that the source emission is confined to a thin ring of fractional width $W$."159 However. we extend the model of GLS by allowing the image interior to the ring to have a relative surface brightness C relerenced to the outer ring (C=0 reduces to the GLS model).," However, we extend the model of GLS by allowing the image interior to the ring to have a relative surface brightness $C$ referenced to the outer ring $C=0$ reduces to the GLS model)."160 This model introduces two additional free parameters., This model introduces two additional free parameters.161" Finally. we consider a non-parametric fit (""direct inversion.) to (he SBP."," Finally, we consider a non-parametric fit (“direct inversion”) to the SBP."162 We divide the image into Nii equalarea annuli; and find (he Iraction /; of the total flix contributed by each bin 7.," We divide the image into $N_{\rm163bin}$ equal-area annuli, and find the fraction $f_i$ of the total flux contributed by each bin $i$ ."164" In this case the magnilication is given by yp=SO.fjr. where j5 is the magnification of annulus i,"," In this case the magnification is given by $\mu=\sum_i f_i \mu_i$, where $\mu_i$ is the magnification of annulus $i$."165" This adds an additional Ny,—1 parameters (due to the constraint that 55;f;=1).", This adds an additional $N_{\rm bin}-1$ parameters (due to the constraint that $\sum_i f_i = 1$ ).166 We find the best-fit solution by minimizing V? with respect to all of these parameters., We find the best-fit solution by minimizing $\chi^2$ with respect to all of these parameters.167" We celine the lo errors on these parameters as the projection of the A\?=1 hvpersurface on the parameter axes. where Vain ds Che mininnim X? for a given model. and dof=104—Ny, is the number of degrees of freedom for the LO4 data points."," We define the $1\sigma$ errors on these parameters as the projection of the $\Delta \chi^2=1$ hypersurface on the parameter axes, where $\chi^2_{\rm min}$ is the minimum $\chi^2$ for a given model, and ${\rm dof}=104-N_{\rm par}$ is the number of degrees of freedom for the 104 data points."168 We normalize NA?by the factor AZ;/dof because we believe that the errors on the data points are Likely unclerestimatecd. resulling in inflated values of 47.," We normalize $\Delta\chi^2$by the factor $\chi^2_{\rm min}/{\rm dof}$ because we believe that the errors on the data points are likely underestimated, resulting in inflated values of $\chi^2$."169 Errors on the fit parameters determined using these inflated values of A7 would be significantly underestimated., Errors on the fit parameters determined using these inflated values of $\chi^2$ would be significantly underestimated.170 Figure 1 shows the best fit for the double power-law model with no lensing., Figure 1 shows the best fit for the double power-law model with no lensing.171 The parameters and lo errors for (04.02.5.4/4) ave tabulated in Table 1.," The best-fit parameters and $1\sigma$ errors for $(\alpha_1,172\alpha_2, s, \tb)$ are tabulated in Table 1."173 The fit is poor: \?=240.7 for 93 dol., The fit is poor: $\chi^2=240.7$ for $93$ dof.174 Residuals [rom the broken power-law model are shown in Figure 3: the svstematic deviations Irom (his model are apparent., Residuals from the broken power-law model are shown in Figure 3; the systematic deviations from this model are apparent.175 When microlensing is included. (he fit improves cousicerably.," When microlensing is included, the fit improves considerably."176For the extended. GLSmodel. we [ind that the data is best explained by emission solely [roii an outerring of relatively small fractional width.,"For the extended GLSmodel, we find that the data is best explained by emission solely from an outerring of relatively small fractional width."177 Specilicallv. we find best-fit parameters. Wo=0.13. (ui. C—0'? [or f;—0 and Wo=0.16 mis. C=09 [or f= 2.," Specifically, we find best-fit parameters, $W=0.13^{+0.04}_{-0.07}$ , $C=0^{+0.02}$ for $k=0$ and $W=0.16_{-0.08}^{+0.03}$ , $C=0^{+0.03}$ for $k=2$ ."178 The remaining parameters, The remaining parameters179Tore than four decades ago. ? performed the first numerical calculations of the collapse of a molecular cloud core to stellar core formation and beyond.,"More than four decades ago, \cite{Larson1969} performed the first numerical calculations of the collapse of a molecular cloud core to stellar core formation and beyond."180 These one-dimensional radiation wdrodynamical calculations revealed the main stages of protostar formation: an almost isothermal collapse until the inner regions become optically thick. the almost adiabatic formation of the first iydrostatie core (typical radius z5 AU and initial mass z5 Mj). he growth of this core as it accreted from the infalling envelope. he second collapse within this core triggered by the dissociation of molecular hydrogen. the formation of the stellar core (initial radius =2 R. and mass z1.5 My). and. lastly. the long accretion phase of the stellar core to its tinal mass.," These one-dimensional radiation hydrodynamical calculations revealed the main stages of protostar formation: an almost isothermal collapse until the inner regions become optically thick, the almost adiabatic formation of the first hydrostatic core (typical radius $\approx 5$ AU and initial mass $\approx 5$ $_{\rm J}$ ), the growth of this core as it accreted from the infalling envelope, the second collapse within this core triggered by the dissociation of molecular hydrogen, the formation of the stellar core (initial radius $\approx 2$ $_\odot$ and mass $\approx 1.5$ $_{\rm J}$ ), and, lastly, the long accretion phase of the stellar core to its final mass."181 Subsequent one-dimensional (e.g.2). and two-dimensional (22) calculations have not changed this qualitative picture substantially. although the latter have allowed the disc-like structure of rotating first cores to be studied.," Subsequent one-dimensional \citep[e.g.][]{MasInu2000}182 and two-dimensional \citep{Tscharnuter1987, Tscharnuteretal2009} calculations have not changed this qualitative picture substantially, although the latter have allowed the disc-like structure of rotating first cores to be studied."183 The first three-dimensional hydrodynamical calculations to follow the collapse to stellar core formation were performed more than a decade ago by 2.., The first three-dimensional hydrodynamical calculations to follow the collapse to stellar core formation were performed more than a decade ago by \cite{Bate1998}.184 These caleulations of rotating molecular cloud cores examined the non-axisymmetrie evolution of the first core and the second collapse phase., These calculations of rotating molecular cloud cores examined the non-axisymmetric evolution of the first core and the second collapse phase.185 If the first core was rotating rapidly enough it was found to be dynamically unstable to a bar- leading to the formation of trailing spiral arms., If the first core was rotating rapidly enough it was found to be dynamically unstable to a bar-mode leading to the formation of trailing spiral arms.186 Gravitational torques removed angular momentum and rotational support from the inner regions of the first core. quickening the onset of the second collapse and preventing fragmentation during the second collapse phase to form close binaries.," Gravitational torques removed angular momentum and rotational support from the inner regions of the first core, quickening the onset of the second collapse and preventing fragmentation during the second collapse phase to form close binaries."187 Several subsequent studies have investigated this phenomenon in more detail (222).. some also including magnetic fields and finding outflows (?2)..," Several subsequent studies have investigated this phenomenon in more detail \citep*{SaiTom2006, SaiTomMat2008, MacInuMat2010}, some also including magnetic fields and finding outflows \citep{Machidaetal2005, MacInuMat2006}."188 However. all these calculations used barotropic equations of state rather than solve the radiation hydrodynamical problem.," However, all these calculations used barotropic equations of state rather than solve the radiation hydrodynamical problem."189 The first three-dimensional calculations including radiative transfer. that followed collapse to the point of stellar core formation (but not beyond) were ?.. using the flux-limited diffusion approximation. and ?.. using a radiative cooling approximation.," The first three-dimensional calculations including radiative transfer that followed collapse to the point of stellar core formation (but not beyond) were \citet{WhiBat2006}, using the flux-limited diffusion approximation, and \cite{Stamatellosetal2007}, , using a radiative cooling approximation."190 Most recently. radiation magnetohydrodynamical calculations of cloud collapse have been performed (2).. but were stopped before the onset of the second collapse.," Most recently, radiation magnetohydrodynamical calculations of cloud collapse have been performed \citep{Tomidaetal2010}, but were stopped before the onset of the second collapse."191 In this paper. we report results from the first three-dimensional radiation hydrodynamical calculations to follow the collapse of rotating molecular cloud coresbevond the formation of the stellar core.," In this paper, we report results from the first three-dimensional radiation hydrodynamical calculations to follow the collapse of rotating molecular cloud cores the formation of the stellar core."192 We tind that the use of radiation hydrodynamics rather than a barotropic equation of state has little effect up until the formation of the stellar core., We find that the use of radiation hydrodynamics rather than a barotropic equation of state has little effect up until the formation of the stellar core.193 However. with radiative transfer. the energy released by the formation of the stellar core has a dramatic effect on the surrounding dise and envelope and launches a temporaryoutflow field.," However, with radiative transfer, the energy released by the formation of the stellar core has a dramatic effect on the surrounding disc and envelope and launches a temporaryoutflow ."194More than 400. extra-solar. planets. have been. ciscovered orbiting5 solar- aiancl later-typeζNIE stars.E but very little is known,"More than 400 extra-solar planets have been discovered orbiting solar- and later-type stars, but very little is known"195In the previous sections we have concentrated on. broad spectral features.,In the previous sections we have concentrated on broad spectral features.196 Here we examine individual line variations over the course of the orbital period., Here we examine individual line variations over the course of the orbital period.197 Perhaps the most interesting feature is the interval immediately. before. the eclipse (ó 70.90) which coincides with the most. prominent dip seen in soft N-ravs (Buckley et al 1998b)., Perhaps the most interesting feature is the interval immediately before the eclipse $\phi\sim$ 0.90) which coincides with the most prominent dip seen in soft X-rays (Buckley et al 1998b).198 Figure 3 shows spectra starting at ó-0.85 where the spectrum has strong Balmer and Le LE line emission and is typical of a Polar., Figure \ref{dip} shows spectra starting at $\phi$ =0.85 where the spectrum has strong Balmer and He II line emission and is typical of a Polar.199 However. at O=O0.87 the emission lines start to become less prominent. especially towards blucr wavelengths.," However, at $\phi$ =0.87 the emission lines start to become less prominent, especially towards bluer wavelengths."200 At Ó-—0.89 only 5 emissionlines are present Ηο A 4100. 15. Hel A 4388. Hell A 4686. 1h? and Ho.," At $\phi$ =0.89 only 5 emissionlines are present– HeII $\lambda$ 4100, $\delta$, HeI $\lambda$ 4388, HeII $\lambda$ 4686, $\beta$ and $\alpha$."201 Moving later in phase. at ó=0.90 and 0.92. we find that Hell A 4686 is still in emission. while La and. Le? appear in absorption emission.," Moving later in phase, at $\phi=0.90$ and 0.92, we find that HeII $\lambda$ 4686 is still in emission, while $\alpha$ and $\beta$ appear in absorption emission."202 At thisphase the emission. ancl absorbed. components of the Ho line have red shifted radial velocities of ~270 km Land —e8TO km * respectively., At thisphase the emission and absorbed components of the $\alpha$ line have red shifted radial velocities of $\sim$ 270 km $^{-1}$ and $\sim$ 870 km $^{-1}$ respectively.203 Later in phase. at 6=0.94 ancl 0.96. very broad emission lines are seen shortwares of 5000A.. although Hell A 4686 is às narrow as before.," Later in phase, at $\phi$ =0.94 and 0.96, very broad emission lines are seen shortwards of 5000, although HeII $\lambda$ 4686 is as narrow as before."204 In the spectrum taken immecdiately before eclipse. Hel A 5876 is back in absorption and all other lines apart [rom Hell A 4686 are very weak.," In the spectrum taken immediately before eclipse, HeI $\lambda$ 5876 is back in absorption and all other lines apart from HeII $\lambda$ 4686 are very weak."205 By ὁ~0.05. the narrow component of the lines has returned.," By $\phi\sim$ 0.05, the narrow component of the lines has returned."206 La 86 we discuss the interpretation of these dip spectra while in the next section we examine the line profile variations over the whole of the orbital period in more detail., In 6 we discuss the interpretation of these dip spectra while in the next section we examine the line profile variations over the whole of the orbital period in more detail.207 We fitted the emission lines using a single Caussian profile for the Helium lines and two Gaussian profiles in the case of the Lyvdrogen lines (there were some spectra where an aciditional Gaussian improved the fit to the Le lines. but not in any consistent manner).," We fitted the emission lines using a single Gaussian profile for the Helium lines and two Gaussian profiles in the case of the Hydrogen lines (there were some spectra where an additional Gaussian improved the fit to the He lines, but not in any consistent manner)."208 The line-Dux was determined by simply summing the flux under the Gaussian(s)., The line-flux was determined by simply summing the flux under the Gaussian(s).209 The flux variations of Ho. IL2. Heb (5876) and Hell (4686) are shown in the top panel of figure 4.. (," The line-flux variations of $\alpha$, $\beta$, HeI (5876) and HeII (4686) are shown in the top panel of figure \ref{rvew}. ("210We only show the results from the night of Feb 8/9 since on the previous night the sampling was lower).,We only show the results from the night of Feb 8/9 since on the previous night the sampling was lower).211 The radial velocities and EW for the narrow ancl broad. components of Ho. I3. Hel(5876) and Holl(4686) are shown in figure 4..," The radial velocities and EW for the narrow and broad components of $\alpha$ , $\beta$ , HeI(5876) and HeII(4686) are shown in figure \ref{rvew}. ."212 The narrow ancl broad. components were fitted with a sinusoidal function of the form: (οτι A, The narrow and broad components were fitted with a sinusoidal function of the form: )= + K213Study of X-ray sources 1ji external galaxies is nuportau for understanding the formation listory and poptlation statistics ο© N-ray binaries axd other N-rav sources both iu external galaxies axd in our own.,Study of X-ray sources in external galaxies is important for understanding the formation history and population statistics of X-ray binaries and other X-ray sources both in external galaxies and in our own.214 Such stiies will help us uuderstaxd the evolulonary history of X-ray biuaries. shotld provide information oi the star formation history WWhite 6josh 1998). and nav be Huportant i estimaine the rate of imereine objects BBethe Brown 1999) critical for determing the rate ¢of gravitational wave events.," Such studies will help us understand the evolutionary history of X-ray binaries, should provide information on the star formation history White Ghosh 1998), and may be important in estimating the rate of merging objects Bethe Brown 1999) critical for determining the rate of gravitational wave events."215EN The nearby. brigh spiral galaxy ΑΟ. offers an excellent site for such studies.," The nearby, bright spiral galaxy M31 offers an excellent site for such studies."216 The distauce of the galaxw is well known. making luinesitv estimation frou flux measurement straight forward. the galaxw is sutiicicutly inclined so that sources can be reliably located within its morpholosx. and it is relatively uearby (780 kpc) so that lüanv sources eau Ὁe detected.," The distance of the galaxy is well known, making luminosity estimation from flux measurement straight forward, the galaxy is sufficiently inclined so that sources can be reliably located within its morphology, and it is relatively nearby (780 kpc) so that many sources can be detected."217 The sub-ucsecond resolution of the Chandra N-rav Observatory (CXO: Weisskopf 1sS) permits individual N-rav sources to be disceried even in crowded reeious of the ealaxy aud enables measurement of hiehlv accurate positions., The sub-arcsecond resolution of the Chandra X-ray Observatory (CXO; Weisskopf 1988) permits individual X-ray sources to be discerned even in crowded regions of the galaxy and enables measurement of highly accurate positions.218 Such position information is critical in fiudiis unique optical and radio counterparts to the SOULCOR., Such position information is critical in finding unique optical and radio counterparts to the sources.219 Tere. we repor on a deep observation of the core of M31 τιmade with the Chandra Ilüeh Resolution Camera (WRC: Murray et 11997).," Here, we report on a deep observation of the core of M31 made with the Chandra High Resolution Camera (HRC; Murray et 1997)."220 The URC offers the vost spatial resolution available for the N-arav stuv of M31., The HRC offers the best spatial resolution available for the X-ray study of M31.221 We describe the, We describe the222"from the average spacing of all unstable radial modes, while f corresponds to the frequency of the lowest unstable radial mode in the interpolated model.","from the average spacing of all unstable radial modes, while $f$ corresponds to the frequency of the lowest unstable radial mode in the interpolated model."223" Finally, the s and f values from different stellar models with the same logg value were averaged to obtain the grid points."," Finally, the $s$ and $f$ values from different stellar models with the same $\log g$ value were averaged to obtain the grid points."224" For this step, models cooler than the theoretical Red Edge given by Dupret et al. ("," For this step, models cooler than the theoretical Red Edge given by Dupret et al. ("2252004) were rejected.,2004) were rejected.226" The observed positions derived from the centers of the clusters are shown for FG Vir, 44 Tau and BL Cam."," The observed positions derived from the centers of the clusters are shown for FG Vir, 44 Tau and BL Cam."227 The grid also provides information about the order of the radial mode that corresponds to the first cluster in the observed frequency spectrum., The grid also provides information about the order of the radial mode that corresponds to the first cluster in the observed frequency spectrum.228 The s-f diagram is in some respect similar to Petersen diagrams (period ratios of consecutive overtones versus the longer period of each pair)., The s-f diagram is in some respect similar to Petersen diagrams (period ratios of consecutive overtones versus the longer period of each pair).229" Petersen diagrams also allow to determine the order of observed radial modes (see, e.g., Olech et al."," Petersen diagrams also allow to determine the order of observed radial modes (see, e.g., Olech et al."230" 2005, Fig."," 2005, Fig."231 6 there)., 6 there).232 A transition between main-sequence models and sequence models takes place at logg values between 3.90 and 3.75., A transition between main-sequence models and sequence models takes place at $\log g$ values between 3.90 and 3.75.233 During the main-sequence stage the logg value decreases due to the slow expansion of the star., During the main-sequence stage the $\log g$ value decreases due to the slow expansion of the star.234" After the TAMS, the star contracts and its temperature and logg values increase again until the second turning point is reached; then the effective temperature and logg start to decrease again."," After the TAMS, the star contracts and its temperature and $\log g$ values increase again until the second turning point is reached; then the effective temperature and $\log g$ start to decrease again."235" For this reason, during the evolution of a star a logg value of 3.80, for example, can be reached three times."," For this reason, during the evolution of a star a $\log g$ value of 3.80, for example, can be reached three times."236 Due to the different stellar structure the radial frequency separation is slightly different., Due to the different stellar structure the radial frequency separation is slightly different.237" Therefore, for the logg value of 3.80, three horizontal grid lines (instead of one) are shown."," Therefore, for the $\log g$ value of 3.80, three horizontal grid lines (instead of one) are shown."238 The lowest of the three grid lines represents main-sequence models and the uppermost grid line post-main sequence models after the second turning point in the HRD., The lowest of the three grid lines represents main-sequence models and the uppermost grid line post-main sequence models after the second turning point in the HRD.239 We will now examine the uncertainties of the grid in detail., We will now examine the uncertainties of the grid in detail.240" In the HRD, models with the same logg values are located almost parallel to the ZAMS."," In the HRD, models with the same $\log g$ values are located almost parallel to the ZAMS."241 Models with different masses and effective temperatures were used to obtain average values of s and f., Models with different masses and effective temperatures were used to obtain average values of $s$ and $f$.242" As can be seen in the HRD in Fig. 9,,"," As can be seen in the HRD in Fig. \ref{fig:hrd_instab},"243" the models, for which the fundamental mode is predicted to be the lowest unstable radial mode, span the largest temperature and mass range."," the models, for which the fundamental mode is predicted to be the lowest unstable radial mode, span the largest temperature and mass range."244 The first overtone is the predicted lowest unstable radial mode for models between the blue edge of the first radial overtone and the blue edge of the radial fundamental mode (only models located inside this region were used to compute the grid points marked as 1H)., The first overtone is the predicted lowest unstable radial mode for models between the blue edge of the first radial overtone and the blue edge of the radial fundamental mode (only models located inside this region were used to compute the grid points marked as 1H).245" Therefore, the uncertainties in the case of the radial fundamental are larger than for the cases in which the lowest unstable radial mode is of higher order."," Therefore, the uncertainties in the case of the radial fundamental are larger than for the cases in which the lowest unstable radial mode is of higher order."246" In Fig. 10,,"," In Fig. \ref{fig:sfgriderr},"247 these uncertainties are given., these uncertainties are given.248 The uncertainties of the grid in logg amount to < 0.05., The uncertainties of the grid in $\log g$ amount to $\leq$ 0.05.249 The highest uncertainties exist for stars in the transition zone between the main sequence and, The highest uncertainties exist for stars in the transition zone between the main sequence and250There are three length scales discussed in connection with GRBs.,There are three length scales discussed in connection with GRBs.251 The first is the size of the central engine (2y~LO°em [or compact objects. Ry~LOtem for collapsars).," The first is the size of the central engine $R_0 \sim 10^6{\rm cm}$ for compact objects, $R_0\sim 10^{11}{\rm cm}$ for collapsars)."252 This region has been the locus of many detailed attempts at understanding the dvnamics of the collapsing object giving rise to the relativistic flow., This region has been the focus of many detailed attempts at understanding the dynamics of the collapsing object giving rise to the relativistic flow.253 The second length scale is the scale associated with the shocking that generates the observed signal. VH for internal shocks or ~LOMem for external shocks.," The second length scale is the scale associated with the shocking that generates the observed signal, $\gamma_f ^2 R_0$ for internal shocks or $\sim 10^{15}{\rm cm}$ for external shocks."254 For obvious reasons. this scale also has been the subject of many studies.," For obvious reasons, this scale also has been the subject of many studies."255 In (his paper we have considered the third. intermediate scale (5405) associated with the acceleration of the thermal fireball.," In this paper we have considered the third, intermediate scale $\sim \gamma R_0$ ) associated with the acceleration of the thermal fireball."256 This region is rich in nuclear and parlicle phenomena and is the only region sensitive (ο. e.g.. the neutron (o proton ratio in ihe flow.," This region is rich in nuclear and particle phenomena and is the only region sensitive to, e.g., the neutron to proton ratio in the flow."257 We have explored nucleosvuthesis for the broad range of dvnanmic timescales. neutroΕν to proton ratios. aud entropy per barvon which may be found in these relativistic [low:JW. surrounding stellar endpoint events.," We have explored nucleosynthesis for the broad range of dynamic timescales, neutron to proton ratios, and entropy per baryon which may be found in these relativistic flows surrounding stellar endpoint events."258 For relativistic flows for which the neutrons remain kinematically well coupled. the nucleosvuthesis is similar (o the lreeze-out uucleosvuthesiV. discussed in connection with BBN.," For relativistic flows for which the neutrons remain kinematically well coupled, the nucleosynthesis is similar to the freeze-out nucleosynthesis discussed in connection with BBN."259 The final mass fraction of deuterium can be greater than a lew percent. some three orders of magnitude larger (han the primordial value.," The final mass fraction of deuterium can be greater than a few percent, some three orders of magnitude larger than the primordial value."260 For flows for which (he assumption of kinematically well coupled neutrons breaks down. light element svnthesis is dominated by high energy non-thermal nucleon-nucleus collisions.," For flows for which the assumption of kinematically well coupled neutrons breaks down, light element synthesis is dominated by high energy non-thermal nucleon-nucleus collisions."261 In these latter cases (he deuterium number fraction can be high. Yp<10%.," In these latter cases the deuterium number fraction can be high, $Y_D\lesssim 10\%$."262 This local overabundance of deuterium will reside in (he shell propagating into the ISM at late times., This local overabundance of deuterium will reside in the shell propagating into the ISM at late times.263 Recently. these shells have gained some attention because a radio identification of à voung (~ LOO vr) IU shell might tell us something about GRBs and their association wilh supernovae (Paczvuski2001).," Recently, these shells have gained some attention because a radio identification of a young $\sim$ 100 yr) non-spherical HI shell might tell us something about GRBs and their association with supernovae \citep{pacz}."264. Whether or not the deuterium produced in stellar endpoint events is detectable depends ultimatelv on the degree of mixing with the ISM., Whether or not the deuterium produced in stellar endpoint events is detectable depends ultimately on the degree of mixing with the ISM.265 The total amount of mass in deuterium produced in these events is Mp2LOο)(Y7/10.2). where E is the total energy in the relativistic flow.," The total amount of mass in deuterium produced in these events is ${\rm M_D} \approx 10^{-6} {\rm M}_{\odot} (E/10^{52}266erg)(100/\gamma_f)(Y_d/10^{-2})$, where $E$ is the total energy in the relativistic flow."267 For IH column densities of order 107?/eni?. this deuteriim will lead to a [actor of (wo increase in ?IL/IE over the primordial ?IL/IT as long as it mixes in a volume less than about a cubic light vear.," For HI column densities of order $10^{19}/{\rm268cm^2}$, this deuterium will lead to a factor of two increase in $\hh$ /H over the primordial $\hh$ /H as long as it mixes in a volume less than about a cubic light year."269 Structures of this size associated with nearby supernova remnants are seen and resolved with the VLA., Structures of this size associated with nearby supernova remnants are seen and resolved with the VLA.270 However. even with the most optimistic assumption that every supernova is accompanied by a relativistic wind. the mechanism discussed here can only contribute to theoverall present day deuterium abundance al (he level of about 0.1 percent.," However, even with the most optimistic assumption that every supernova is accompanied by a relativistic wind, the mechanism discussed here can only contribute to the present day deuterium abundance at the level of about 0.1 percent."271 It is perhaps frustrating then that such a prodigious vield of deuterium has little leverage on the scale of the cosmological abundance., It is perhaps frustrating then that such a prodigious yield of deuterium has little leverage on the scale of the cosmological abundance.272 However. the potentially large ?II production in," However, the potentially large $\hh$ production in"273the sound velocity. which ts significantly lower in the core than expected from simple extrapolation below the outer layers.,"the sound velocity, which is significantly lower in the core than expected from simple extrapolation below the outer layers."274 In this case. the sign inversion in the integral is sufficient to explain why the small separations become negative (Fig. 4)).," In this case, the sign inversion in the integral is sufficient to explain why the small separations become negative (Fig. \ref{fig4}) )."275 The frequency for which this occurs corresponds to that of the {=2 waves for which the turning point is at the core boundary., The frequency for which this occurs corresponds to that of the $\ell=2$ waves for which the turning point is at the core boundary.276 In other cases. it may happen that the integral value changes abruptly at the core boundary. but not enough to become positive.," In other cases, it may happen that the integral value changes abruptly at the core boundary, but not enough to become positive."277 Then. a slight difference between the large separations for the /= 0and the £=2 modes may be enough to lead to the inversion of the sign of the small separations (Fig. 3)).," Then, a slight difference between the large separations for the $\ell=0$ and the $\ell=2$ modes may be enough to lead to the inversion of the sign of the small separations (Fig. \ref{fig3}) )."278 In this case. the frequency where this behaviour occurs may be slightly different from that related to the core boundary.," In this case, the frequency where this behaviour occurs may be slightly different from that related to the core boundary."279 As a conclusion. that the small separations change sign is related to the core boundary and may be physically understood.," As a conclusion, that the small separations change sign is related to the core boundary and may be physically understood."280 However the use of this sign inversion to derive the core size can only be done with precise model computations., However the use of this sign inversion to derive the core size can only be done with precise model computations.281 Figures 5 to 7. present the internal parameters for three models along an evolutionary track. of 1.15 M..with [Fe/H]=0.30. without overshooting.," Figures \ref{fig5} to \ref{fig7} present the internal parameters for three models along an evolutionary track of 1.15 with [Fe/H]=0.30, without overshooting."282 The first model lies on the main sequence. in a situation where the small separations are always positive in the considered frequency range (1.782 Gyr. Fig. 5)).," The first model lies on the main sequence, in a situation where the small separations are always positive in the considered frequency range (1.782 Gyr, Fig. \ref{fig5}) )."283 The second model is more evolved. and the small separations for the degrees £=0 - 2 become negative at a frequency of 3.4 mHz (4.654 Gyr. Fig.6)).," The second model is more evolved, and the small separations for the degrees $\ell=0$ - $\ell=2$ become negative at a frequency of 3.4 mHz (4.654 Gyr, \ref{fig6}) )."284" It corresponds to what we have called the “transition model"".", It corresponds to what we have called the “transition model”.285 The third model is still more evolved on the subgiant branch (5.937 Gyr. Fig. 7)).," The third model is still more evolved on the subgiant branch (5.937 Gyr, Fig. \ref{fig7}) )."286 In each figure. four different graphs are presented: the small separations (upper-left panel). the echelle diagram (upper-right panel. the sound speed profile (lower-left panel). and the integral 7607) (lower-right panel).," In each figure, four different graphs are presented: the small separations (upper-left panel), the echelle diagram (upper-right panel), the sound speed profile (lower-left panel), and the integral $I(r)$ (lower-right panel)."287 At the bee!ning of the main sequence (model presented in Fig.5)). the covective core of the star is still small (ΔΝ 0.03).," At the beginning of the main sequence (model presented in \ref{fig5}) ), the convective core of the star is still small $R_{cc}/R=0.03$ )."288 The certral helium abundance is low and there is no discontinuity 1 the sound speed profile at the centre of the star., The central helium abundance is low and there is no discontinuity in the sound speed profile at the centre of the star.289 The integeral /(r) is negative in the whole star., The integral $I(r)$ is negative in the whole star.290 In that case. the small separations 9vo» are positive for the full range of frequencies considered. and there is no crossing point in the echelle diagram.," In that case, the small separations $\delta\nu_{02}$ are positive for the full range of frequencies considered, and there is no crossing point in the echelle diagram."291 When the age of the star increases. the values of the small separations decrease.," When the age of the star increases, the values of the small separations decrease."292 The convective core develops. and there is more helium in the centre of the star.," The convective core develops, and there is more helium in the centre of the star."293 A discontinuity in the chemical composition and in the sound speed profile appears., A discontinuity in the chemical composition and in the sound speed profile appears.294 The model shown in Fig., The model shown in Fig.295 6 is the first model for which we found negative small separations below 3.5 mHz., \ref{fig6} is the first model for which we found negative small separations below 3.5 mHz.296" We call it the “transition model"" and show its position as a cross on the", We call it the “transition model” and show its position as a cross on the297(M<1M..) ἔτι) (~ Lega). ,$M < 1M_{\rm \odot}$ $L_{\mathrm{Edd}}$ $\sim$ $L_{\mathrm{Edd}}$ 298The images of an arbitrary source near a cusp can be discussed just as easily.,The images of an arbitrary source near a cusp can be discussed just as easily.299 In the lowest order approximation. where u(7. the lens equation (26)) near a cusp reads as follows.," In the lowest order approximation, where $u \sim v^2$, the lens equation \ref{eqCusp}) ) near a cusp reads as follows."300" b= 2U A,", x = 2 u - _2 v^2; y = - v _2 2u - _3 v^2)301motion.,motion.302 This was observed in 47 Tuc W (Bogdanovetal.2005) which exhibits large-amplitude X-ray variability. probably due to geometric occultations of an X-ray emitting intra-binary shock by the companion main-sequence star (Bogdanovetal.2006).," This was observed in 47 Tuc W \citep{Bogdanov2005} which exhibits large-amplitude X-ray variability, probably due to geometric occultations of an X-ray emitting intra-binary shock by the companion main-sequence star \citep{Bogdanov2006}."303. As an example. Figure 14. shows the lightcurves of the M28 pulsars PSR J1824-2452G and J1824-2452H as folded at their binary period.," As an example, Figure \ref{figure14} shows the lightcurves of the M28 pulsars PSR J1824-2452G and J1824-2452H as folded at their binary period."304 PSR J1824-2452G is in a hh binary system with a low-mass companion., PSR J1824-2452G is in a h binary system with a low-mass companion.305 J1824-2452H has an orbit period of hh and has been observed to show radio eclipses (Bégin2000)., J1824-2452H has an orbit period of h and has been observed to show radio eclipses \citep{Begin2006}.306". Indeed, inspecting the lighteurves by eye may reveal a flux variability with the pulsar’s orbital binary motion."," Indeed, inspecting the lightcurves by eye may reveal a flux variability with the pulsar's orbital binary motion."307" From the statistical point of view, though, the significance for variability depends on the number of phase bins and is only at the —1.5—2o level for lightcurves with 6, 8 or 10 phase bins."," From the statistical point of view, though, the significance for variability depends on the number of phase bins and is only at the $\sim 1.5-2\sigma$ level for lightcurves with 6, 8 or 10 phase bins."308 More stringent results were found for PSR J1748-2021B in NGC 6440 and for PSR in NGC 6397., More stringent results were found for PSR J1748-2021B in NGC 6440 and for PSR J1740-5340 in NGC 6397.309 PSR J1748-2021B has an orbital period of ~20 days (Freireetal...2008)., PSR J1748-2021B has an orbital period of $\sim20$ days \citep{Freire2008}.310" ]t was observed by Chandra twice, in 2000 July 4 and three years later in 2003 June 27."," It was observed by Chandra twice, in 2000 July 4 and three years later in 2003 June 27."311 These observations cover the relatively narrow orbital phase 0.124—0.137 and 0.058—0.072., These observations cover the relatively narrow orbital phase $0.124-0.137$ and $0.058-0.072$.312" The vignetting corrected net counting rates which we have measured from the pulsar in this two datasets are (1.68£0.28)x107 cts/s and (5.03£0.46)x107 ets/s, respectively."," The vignetting corrected net counting rates which we have measured from the pulsar in this two datasets are $(1.68\pm 0.28) \times10^{-3}$ cts/s and $(5.03\pm0.46)\times10^{-3}$ cts/s, respectively."313 The counting rates measured in both data thus differ by a factor of ~3 with a significance of ~66.," The counting rates measured in both data thus differ by a factor of $\sim3$ with a significance of $\sim 6\,\sigma$."314 On whether this variability is because of a modulation of the X-ray flux over the pulsar's binary orbit or because of a long term flux increase by other means can not be clarified with the available data., On whether this variability is because of a modulation of the X-ray flux over the pulsar's binary orbit or because of a long term flux increase by other means can not be clarified with the available data.315 The millisecond pulsar PSR J1740-5340 in NGC 6397 is in à ~1.35 day binary orbit with a massive late type companion., The millisecond pulsar PSR J1740-5340 in NGC 6397 is in a $\sim1.35$ day binary orbit with a massive late type companion.316 The pulsar's radio emission is seen to eclipse in the orbital phase interval 0.05—0.45 (D'Amicoetal...2001)., The pulsar's radio emission is seen to eclipse in the orbital phase interval $0.05-0.45$ \citep{DAmico2001}.317. Five datasets from NGC 6397 are available in the Chandra archive (cf., Five datasets from NGC 6397 are available in the Chandra archive (cf.318 Table 1)). covering various phase ranges of the pulsar's binary orbit (see Huang Becker 2010 for a more detailed discussion).," Table \ref{t:observations}) ), covering various phase ranges of the pulsar's binary orbit (see Huang Becker 2010 for a more detailed discussion)."319" The first observation done in 2000 July 3] was aimed on the front-illuminated (FD) ACIS-13 chip, while the other fourobservations were taken with the back-illuminated (BI) chip ACIS-S3."," The first observation done in 2000 July 31 was aimed on the front-illuminated (FI) ACIS-I3 chip, while the other fourobservations were taken with the back-illuminated (BI) chip ACIS-S3."320" The pulsar's net counting rates obtained from this data are (1.31£0.17)x10? ets/s. (1.7420.25)x107 cts/s, (2.89£0.33)xLOM ets/s, (2.2640.16)x107 cts/s and (2.92£0.14)x107 ets/s. respectively. revealing a 3o flux variability on time scales of month to years."," The pulsar's net counting rates obtained from this data are $(1.31\pm0.17)\times 32110^{-3}$ cts/s, $(1.74\pm 0.25)\times 10^{-3}$ cts/s, $(2.89\pm0.33)\times 10^{-3}$ cts/s, $(2.26\pm0.16)\times 10^{-3}$ cts/s and $(2.92\pm0.14)\times 10^{-3}$ cts/s, respectively, revealing a $\sim 3\sigma$ flux variability on time scales of month to years."322 Figure 15 shows the pulsar lightcurve as folded at the binary period., Figure \ref{figure15} shows the pulsar lightcurve as folded at the binary period.323" The significance of the flux modulation over the observed orbit was found to be between and 99,65c,, depending on the number of phase bins used to construct the lightcurve (cf."," The significance of the flux modulation over the observed orbit was found to be between and , depending on the number of phase bins used to construct the lightcurve (cf."324 Huang Becker 2010)., Huang Becker 2010).325the formation of dense clusters within a cloud supported by turbulence must occur over several cloud crossing times as argued by ? and ?..,the formation of dense clusters within a cloud supported by turbulence must occur over several cloud crossing times as argued by \citet{tan06} and \citet{krumholz06}.326 ? expresses a different view point by proposing that star formation becomes efficient within cores with a mean density >10 H. em? independent of turbulence., \citet{elmegreen08} expresses a different view point by proposing that star formation becomes efficient within cores with a mean density $> 10^4~$ H $^{-3}$ independent of turbulence.327 These ideas are being tested against observations of galactic star forming regions and cores. which are far more detailed than the present information we have on the compact ssource.," These ideas are being tested against observations of galactic star forming regions and cores, which are far more detailed than the present information we have on the compact source."328 High angular resolution observations of molecular lines at mm wavelengths. which unlike the near-IR lines will trace the kinematics of the bulk of the gas and the source density structure. would be needed for a detailed study of the formatior of super star clusters on the example of this source. and if it is consistent with present ideas based on observations of lower mass and less turbulent cores.," High angular resolution observations of molecular lines at mm wavelengths, which unlike the near-IR lines will trace the kinematics of the bulk of the gas and the source density structure, would be needed for a detailed study of the formation of super star clusters on the example of this source, and if it is consistent with present ideas based on observations of lower mass and less turbulent cores."329 Here we only make preliminary remarks that can soon be tested with ALMA., Here we only make preliminary remarks that can soon be tested with ALMA.330" The mean density ofthe compact ssource is ~10 H em"".", The mean density ofthe compact source is $ \sim 10^4~$ H $^{-3}$.331 We estimate its formation time scale to at least 10° yr. but the star-formation efficiency is still very small (stellar to gas mass ratio ~ 0.3%).," We estimate its formation time scale to at least $10^6 \,$ yr, but the star-formation efficiency is still very small (stellar to gas mass ratio $\sim 0.3\%$ )."332 This remarkable result may indicate that the high turbulence (1D velocity dispersion 20kms! ) has been very effective at preventing star formation.," This remarkable result may indicate that the high turbulence (1D velocity dispersion $\rm \sim 20 \, km \, s^{-1}$ ) has been very effective at preventing star formation."333 Star formation may occur rapidly once the rate of mass aceretion becomes insufficient to drive the present amplitude of turbulence., Star formation may occur rapidly once the rate of mass accretion becomes insufficient to drive the present amplitude of turbulence.334 The fact that stars may be forming out of gas which has lost the turbulent energy it had during the initial gravitational contraction reduces the requirement on the star formation efficiency to form a bound stellar cluster., The fact that stars may be forming out of gas which has lost the turbulent energy it had during the initial gravitational contraction reduces the requirement on the star formation efficiency to form a bound stellar cluster.335 The star formation rate (SFR) in the Antennae merger has been estimated to 20 vr! for a total molecular gas mass 10'?M...," The star formation rate (SFR) in the Antennae merger has been estimated to $\rm 20 \, $ $^{-1}$ for a total molecular gas mass $\sim 10^{10} \, $."336 Local values measured from CO and Πα images of the Antennae agree with the general correlation between the SFR and molecular gas surface densities. the Schmidt-Kennicutt law (??)..," Local values measured from CO and $\alpha$ images of the Antennae agree with the general correlation between the SFR and molecular gas surface densities, the Schmidt-Kennicutt law \citep{kennicutt98,zhang01}."337 The burst of star formation observed in the overlap regtor follows mainly from the high surface density of molecular gas with only a small (a factor ~ 3) enhancement of the star formation efficiency with respect to the Galactic value., The burst of star formation observed in the overlap region follows mainly from the high surface density of molecular gas with only a small (a factor $\sim 3$ ) enhancement of the star formation efficiency with respect to the Galactic value.338 This enhancement comes from the non-linear dependence of the SFR on the surface density of gas., This enhancement comes from the non-linear dependence of the SFR on the surface density of gas.339 The star formation rate per unit gas mass in the SINFONI field of view is similar to the global value., The star formation rate per unit gas mass in the SINFONI field of view is similar to the global value.340 Based on their Spitzer and Herschel data. ? and ? find an SPR of ~ 0.7 yyr! for a molecular mass of4x105 M...," Based on their Spitzer and Herschel data, \citet{brandl09} and \cite{klaas10} find an SFR of $\sim$ 0.7 $^{-1}$ for a molecular mass of $4\tten{8}$ ."341 To compute the star formation efficiency. we consider that star formation is triggered by gas compression and occurs over the characteristic time. ~ 1OMyr. over which the tidal interaction is compressive (?)..," To compute the star formation efficiency, we consider that star formation is triggered by gas compression and occurs over the characteristic time, $\sim 10\,$ Myr, over which the tidal interaction is compressive \citep{renaud08}."342 This is also the crossing time across the SINFONI field of view for the CO line width., This is also the crossing time across the SINFONI field of view for the CO line width.343 Over this time the gas mass converted into stars is 7x10?M.... which yields a star formation efficiency of2%..," Over this time the gas mass converted into stars is $\tten{6}$, which yields a star formation efficiency of."344 The SSC accounts for a significant fraction of this stellar mass (see estimates of the cluster mass in Table 1))., The SSC accounts for a significant fraction of this stellar mass (see estimates of the cluster mass in Table \ref{tab:sc}) ).345 For the SGMC 2 complex. like for the compact ssource. the low star-formation efficiency may result from the strong. driven turbulence.," For the SGMC 2 complex, like for the compact source, the low star-formation efficiency may result from the strong, driven turbulence."346 For the molecular complex. the star-formation efficiency may remain low until it is disrupted by the tidal interaction.," For the molecular complex, the star-formation efficiency may remain low until it is disrupted by the tidal interaction."347 If turbulence continues to be driven by on-going accretion. it is possible that the CO complex will be dispersed without losing its turbulent energy.," If turbulence continues to be driven by on-going accretion, it is possible that the CO complex will be dispersed without losing its turbulent energy."348 It will be interesting to extend the present study to other complexes in the overlap region to test this idea., It will be interesting to extend the present study to other complexes in the overlap region to test this idea.349 It is interesting to compare this analysis and interpretation of observational data with numerical simulations., It is interesting to compare this analysis and interpretation of observational data with numerical simulations.350 In. their Figure |. ? compare the star-formation rates in recent numerical simulations of the Antennae.," In their Figure 1, \citet{karl10b} compare the star-formation rates in recent numerical simulations of the Antennae."351 All simulations predict a significant enhancement of the star-formation efficiency. relative to its value in the two spirals prior to interaction. after pericenter passage.," All simulations predict a significant enhancement of the star-formation efficiency, relative to its value in the two spirals prior to interaction, after pericenter passage."352 In the simulations. the enhancement in the star formation efficiency is triggered by the gas compression and subsequent gas cooling.," In the simulations, the enhancement in the star formation efficiency is triggered by the gas compression and subsequent gas cooling."353 For several runs the enhancement is one order of magnitude or more. 1.8. larger than that derived from observations.," For several runs the enhancement is one order of magnitude or more, i.e. larger than that derived from observations."354 ? present new simulations that quantify the impact of stellar feedback on the star formation efficiency. and argue that it is necessary to invoke stellar feedback to compensate for the gas cooling.," \citet{karl10b} present new simulations that quantify the impact of stellar feedback on the star formation efficiency, and argue that it is necessary to invoke stellar feedback to compensate for the gas cooling."355 Our oobservations do not support this idea., Our observations do not support this idea.356 In Sect. ??..," In Sect. \ref{sec:large},"357 we argue that the gas turbulence is driven by the galaxy interaction., we argue that the gas turbulence is driven by the galaxy interaction.358 Stellar feedback is energetically significant but not a key contributor to the turbulent energy of the molecular gas., Stellar feedback is energetically significant but not a key contributor to the turbulent energy of the molecular gas.359 This holds for the SGMC 2 complex and the compact ssource., This holds for the SGMC 2 complex and the compact source.360 We start discussing feedback on the scale of the SGMC 2 complex using Starburst99 models., We start discussing feedback on the scale of the SGMC 2 complex using Starburst99 models.361 For a Salpeter IMF and continuous star formation over 10 Myr. the mechanical energy associated with stellar winds and supernovae explosions is Lytecn(SF)~4x10° (SER/0.7 yyr!)Ls.," For a Salpeter IMF and continuous star formation over $10~$ Myr, the mechanical energy associated with stellar winds and supernovae explosions is $L_\mathrm{Mech}(\mathrm{SF}) \sim 4\times36210^{7} \times$ (SFR/0.7 $^{-1})~L_\odot$."363 Most of this energy is released by the stars in the SSC., Most of this energy is released by the stars in the SSC.364 The non-thermal component of the radio flux from the SSC (?) indicates that the SSC has evolved to an age where the most massive stars are exploding as supernovae., The non-thermal component of the radio flux from the SSC \citep{neff00} indicates that the SSC has evolved to an age where the most massive stars are exploding as supernovae.365 For a stellar mass of a few 10°M... the mechanical power from stellar winds and supernovae is also a few 10 L...," For a stellar mass of a few $10^6 \, $, the mechanical power from stellar winds and supernovae is also a few $\rm36610^7$ $L_\odot$."367 This value is one order of magnitude larger than the luminosity of the extendec emission. but the lack of enhancement of the line width towards the SSC is evidence that stellar feedback does not contribute significantly to driving the turbulent kinetic energy of the geas.," This value is one order of magnitude larger than the luminosity of the extended emission, but the lack of enhancement of the line width towards the SSC is evidence that stellar feedback does not contribute significantly to driving the turbulent kinetic energy of the gas."368 The energy from stellar feedback must be mostly transferred to the hot X-ray emitting plasma (2).., The energy from stellar feedback must be mostly transferred to the hot X-ray emitting plasma \citep{baldi06}. .369 We plan to test this tentative conclusion. with additional SINFONI data towardsother super star clusters in the overlap region., We plan to test this tentative conclusion with additional SINFONI data towardsother super star clusters in the overlap region.370 For the compact source. we estimate the star formation rate by dividing the stellar mass of4x10! ," For the compact source, we estimate the star formation rate by dividing the stellar mass of $4\times 10^4$ "371 We now reproduce the analysis of Sect., We now reproduce the analysis of Sect.372" 4.3] for the same inputs, but after having added Gaussian noise to the input fluxes (5 %))."," \ref{sec:UBVIK_nonoise_Av} for the same inputs, but after having added Gaussian noise to the input fluxes $5$ )."373 Figure [/| presents the results., Figure \ref{fig:UBVIK_noise} presents the results.374 At the top of this figure are also plotted estimates derived from the single best-y7 fit., At the top of this figure are also plotted estimates derived from the single $\chi^2$ fit.375" Although the inputs are now noisy, the trends obtained remain similar."," Although the inputs are now noisy, the trends obtained remain similar."376" In particular, there is no important difference with the noise-free case when the analysis is based on standard, continuous population synthesis models (bottom panels): most fits are poor, ages tend to be underestimated, masses are highly dispersed around the correct values."," In particular, there is no important difference with the noise-free case when the analysis is based on standard, continuous population synthesis models (bottom panels): most fits are poor, ages tend to be underestimated, masses are highly dispersed around the correct values."377 Bayesian estimates remain close to the expected values even though the dispersion has increased., Bayesian estimates remain close to the expected values even though the dispersion has increased.378 The standard deviation of the residuals are of 0.15 dex in age and 0.13 dex in mass except for models in 6.a and 6.0 features., The standard deviation of the residuals are of $0.15$ dex in age and $0.13$ dex in mass except for models in $6.a$ and $6.b$ features.379 Only 13 of the model clusters are assigned underestimated ages., Only $13$ of the model clusters are assigned underestimated ages.380 A pleasant fact is that the single best-y? fit provides results that are similar to the Bayesian ones[6., A pleasant fact is that the single $\chi^2$ fit provides results that are similar to the Bayesian ones.381"2.1]. Although individual clusters are assigned different ages, the estimated properties of the sample as a whole are described similarly with both methods."," Although individual clusters are assigned different ages, the estimated properties of the sample as a whole are described similarly with both methods."382" For 84 of the clusters, the single best fit age and the Bayesian ages are both within 0.3 dex of the actual ages."," For 84 of the clusters, the single best fit age and the Bayesian ages are both within $0.3$ dex of the actual ages."383" Of the 16 that deviate, 3 deviate only with the single best fit method, 3 only with the full Bayesian method, and 10 deviate in similar ways with both methods."," Of the 16 that deviate, 3 deviate only with the single best fit method, 3 only with the full Bayesian method, and 10 deviate in similar ways with both methods."384"a massive star is «20Myr. and therefore the Tvpe II supernova rate should (race (he star formation rate almost exactly,","a massive star is $<20 \;{\rm Myr}$, and therefore the Type II supernova rate should trace the star formation rate almost exactly."385 For a Salpeter IME. there is one Type II supernova (SN II) per 150M. of star formation. and we consequently divide the SFR. bv this factor (which we call Max(1I) ) to obtain the Pop II ον rate.," For a Salpeter IMF, there is one Type II supernova (SN II) per $\sim 150 \;\msun$ of star formation, and we consequently divide the SFR by this factor (which we call $M_{\rm SN}({\rm II})\;$ ) to obtain the Pop II SN rate."386" Hernquist&Springel(2002) find that the SFR at z«z,, in their simulations actually has a power-law dependence on the expansion rate. rather (han evolving exponentially with redshift: however the fit given in the above equation is sufficient for our present purposes."," \citet{HerSpr02} find that the SFR at $z<z_m$ in their simulations actually has a power-law dependence on the expansion rate, rather than evolving exponentially with redshift; however the fit given in the above equation is sufficient for our present purposes."387 For the PISN rate we use the VAIS star formation rate in Figure 2.., For the PISN rate we use the VMS star formation rate in Figure \ref{fig2}.388 The (vpical lifetime ola VAIS is ~3Myr so we can again assume the SN rate traces the SFR closely.," The typical lifetime of a VMS is $\sim 3 \;{\rm389Myr}$ so we can again assume the SN rate traces the SFR closely."390 Wetake tvpical VMSs to have masses of 250M. ancl assume (μονalf die as supernovae to calculate the VAIS supernova rate Max(1II)=250AL. )., Wetake typical VMSs to have masses of $250 \;\msun$ and assume they die as supernovae to calculate the VMS supernova rate $M_{\rm SN}({\rm III})=250 \;\msun$ ).391 VMSs are predicted to end their lives as pair-inslability supernovae if Af<260M... ancl these should be very energetic and Iumninous evenis.," VMSs are predicted to end their lives as pair-instability supernovae if $M<260 \;\msun$, and these should be very energetic and luminous events."392 Dear in mind. however. (hat the mass Iunction. aid hence end state of Pop LI stars is still somewhat uncertain. making (his supernova rate a crude estimate.," Bear in mind, however, that the mass function, and hence end state of Pop III stars is still somewhat uncertain, making this supernova rate a crude estimate."393 We also calculate the rate of Pop IL5 supernovae. using (he result [rom the last section that the Pop II.5 SFR is directly proportional to the Pop IHE SER. with constant of proportionality ο1.," We also calculate the rate of Pop II.5 supernovae, using the result from the last section that the Pop II.5 SFR is directly proportional to the Pop III SFR, with constant of proportionality $\eta \sim 1$."394 Due to the uncertain nature of the Pop H.5 IME we must make simple assumptions to eel the mass in stars formed per supernova., Due to the uncertain nature of the Pop II.5 IMF we must make simple assumptions to get the mass in stars formed per supernova.395 The characteristic stellar mass was shown to be about LOM... so we assume that hall of the stellar mass goes into stars that can explode.," The characteristic stellar mass was shown to be about $10\;\msun$, so we assume that half of the stellar mass goes into stars that can explode."396 We further assume that the average mass of a Pop IL5 supernova progenitor is ~20M..., We further assume that the average mass of a Pop II.5 supernova progenitor is $\sim 20\;\msun$.397 From (his we find that. Max1.5)~40M. lor Pop IL5 supernovae., From this we find that $M_{\rm SN}({\rm II.5})\sim40 \;\msun$ for Pop II.5 supernovae.398 We expect (hese supernovae to be of Type II. although with very low metallicity.," We expect these supernovae to be of Type II, although with very low metallicity."399" We convert (hese rales per unit volume into rates observed per unit time and per unit redshift interval: llere dV. is a comoving volume element. May is (he mass in stars formed per supernova for the relevant stellar population. f, the fraction of the skv observed. and ο)7, the IIubble length."," We convert these rates per unit volume into rates observed per unit time and per unit redshift interval: Here ${\rm d}V$ is a comoving volume element, $M_{\rm SN}$ is the mass in stars formed per supernova for the relevant stellar population, $f_{sky}$ the fraction of the sky observed, and ${c/H_0}$ the Hubble length."400" Note that the Pop IL5 and Pop III supernova rates differ only in their respective values for M, calculated above.", Note that the Pop II.5 and Pop III supernova rates differ only in their respective values for $M_{\ast}$ calculated above.401" The respective time intervals in the observer ancl emitter frames. d/,4 and d/,,,,. ave related bv the [actor of (1+2) in the denominator."," The respective time intervals in the observer and emitter frames, ${\rm402d}t_{obs}$ and ${\rm d}t_{em}$ , are related by the factor of $(1+z)$ in the denominator."403 The comoving distance (o redshift 2 is, The comoving distance to redshift $z$ is404The RRATs are a group of isolated Galactic neutron stars discovered in an archival search of the Parkes Multi-beam. Pulsar Survey (PAIRS: ?)).,The RRATs are a group of isolated Galactic neutron stars discovered in an archival search of the Parkes Multi-beam Pulsar Survey (PMPS; \citealt{mclaughlin06}) ).405 They are. characterised by milliseconcl racio bursts with [lux σασίies at 1.4 CGllz of 90L) Jv., They are characterised by millisecond radio bursts with flux densities at $1.4$ GHz of $\sim0.1-10$ Jy.406 Phe bursts are infrequent. wit1 intervals of as long as a few hours to as short as a few minutes (2)..," The bursts are infrequent, with intervals of as long as a few hours to as short as a few minutes \citep{keane10}."407 The rotation periods. inferred by dividing the burst intervals by the largest common denominator. lie in the lloT s range.," The rotation periods, inferred by dividing the burst intervals by the largest common denominator, lie in the $0.1-7$ s range."408 Of the 22 RAAT sources in the PMPS with known periods. 1l have periods greater than 4 s. compared ο just 2% of racio pulsars (2)..," Of the 22 RRAT sources in the PMPS with known periods, 11 have periods greater than 4 s, compared to just $\sim2\%$ of radio pulsars \citep{keane10a}."409 Fhese long periods are rerniniscent of the (X-ray cim) isolated neutron stars and the magnetars (see 7? and ?))., These long periods are reminiscent of the (X-ray dim) isolated neutron stars and the magnetars (see \citealt{haberl07} and \citealt{mereghetti08}) ).410 Using their dispersion measure. he RIRATSs are found to lie approximately 17 kpe cistant in the Galactic plane (?)..," Using their dispersion measure, the RRATs are found to lie approximately 1–7 kpc distant in the Galactic plane \citep{keane10a}."411" The nature of RATS is still an open qu«λος,", The nature of RRATs is still an open question.412 A simple estimate of their Galactic population suggest sthat they may be more abundant than the radio pulsars., A simple estimate of their Galactic population suggests that they may be more abundant than the radio pulsars.413 In fact. if we consider RRATSs and the other known classes of neutron stars as clistinet populations. a birthrate problem arises. i.c. neutron stars appear to be formed faster than the observed supernova rate.," In fact, if we consider RRATs and the other known classes of neutron stars as distinct populations, a birthrate problem arises, i.e. neutron stars appear to be formed faster than the observed supernova rate."414 This problem may be resolved. however. if the various neutron star types are instead. regarded: as different evolutionary phases (2)..," This problem may be resolved, however, if the various neutron star types are instead regarded as different evolutionary phases \citep{keane08}."415 Pwo main models have oen proposed for the intermittent pulses. exhibited. by αν with some researchers assigning the phenomenon to detection issues and others favouring intrinsically transient emission.," Two main models have been proposed for the intermittent pulses exhibited by RRATs, with some researchers assigning the phenomenon to detection issues and others favouring intrinsically transient emission."416" ?. suggest that Αν are distant analogues of ""Slt D0656|1H. ie. pulsars with regular emission that is xlow our detection limit but which show large amplitude oulses. drawn from. an extended: pulsc-energy. distribution."," \citet{Weltevrede06}417 suggest that RRATs are distant analogues of PSR B0656+14, i.e. pulsars with regular emission that is below our detection limit but which show large amplitude pulses drawn from an extended pulse-energy distribution."418" Alternatively, 7/— suggest that the bursts in. RIALS are »owered by the episodic injection of material [rom a circumpolar asteroid: belt. formed. from supernova fallback material. that temporarily reactivates a cquiescent region of the magnetosphere."," Alternatively, \citet{cordes08} suggest that the bursts in RRATs are powered by the episodic injection of material from a circumpolar asteroid belt, formed from supernova fallback material, that temporarily reactivates a quiescent region of the magnetosphere."419 Similarly. 7 propose that RRATSs may be surrounded by planctary-like radiation belts. aud the trapped plasma in this belt is episocteally disrupted," Similarly, \citet{luo07} propose that RRATs may be surrounded by planetary-like radiation belts, and the trapped plasma in this belt is episodically disrupted"420We performed a systematic search of known binaries amoue the caucidates of Table 1.,We performed a systematic search of known binaries among the candidates of Table 1.421 HD 35515 and WD 111851 are visual binarics whose speckle interferometric measurements are given by AlcAlister et al (1993)., HD 38545 and HD 141851 are visual binaries whose speckle interferometric measurements are given by McAlister et al (1993).422 Cüven the sinall separation of the pairs. erouud-basecd spectra will always be the combined spectrum of these binaries: for the first star. Iipparcos data demonstrate that the effect of the companion is surely not weeheible.," Given the small separation of the pairs, ground-based spectra will always be the combined spectrum of these binaries; for the first star, Hipparcos data demonstrate that the effect of the companion is surely not negligible."423 Starting from the analysis of the II. profile. Farageiana et al (1997) were able to show hat a thi star. IID 1117586. is indeed a binary svsteni and this finding was supported by the identification of he lines Gu the visual. but not in the UW range below 2000 Aj} of the cooler companion. which are narrow if compared to those of the primary star. this happens quite ikelv because the companion is seeu pole ou.," Starting from the analysis of the $_{\gamma}$ profile, Faraggiana et al (1997) were able to show that a third star, HD 111786, is indeed a binary system and this finding was supported by the identification of the lines (in the visual, but not in the UV range below 2000 ) of the cooler companion, which are narrow if compared to those of the primary star, this happens quite likely because the companion is seen pole on."424 In order to deteruine abundances. the spectra of the two coupoucuts uust be diseutaueled.," In order to determine abundances, the spectra of the two components must be disentangled."425 Twelve ou of SD stars of our snuple are sown to be visual binaries according to he Mermilliod et al. (, Twelve out of 89 stars of our sample are known to be visual binaries according to the Mermilliod et al. (4261997) catalogue.,1997) catalogue.427 For these binaries ouly the combined colour indices have been measured aud therefore they caunot be safelv used to cerive the atmospheric paramcters of the primary componucut., For these binaries only the combined colour indices have been measured and therefore they cannot be safely used to derive the atmospheric parameters of the primary component.428 In order to evaluate the influence of the companion star. we extracted he angular separation and the magnitudes for the components A and D from the Washington Visual Double Star CatakSo (Worley Douglass 1997: hereafter WDS): these data are given iun coluuns 10 and 11 of Table 1 for the objects with an angular separation of less than LO arcsec.," In order to evaluate the influence of the companion star, we extracted the angular separation and the magnitudes for the components A and B from the Washington Visual Double Star Catalog (Worley Douglass 1997; hereafter WDS); these data are given in columns 10 and 11 of Table 1 for the objects with an angular separation of less than 10 arcsec."429 The contamination ou the observed spectra depends ou the huninosity difference of the components aud on the slit width of the spectrograph ou the sky: for several stars the observed spectra are expected to be affected by the companion star and. iu some cases (IID 290192. TD 38515. ΠΟ 17152. ΠΟ 111551. TD 153508. ΠΟ 159082. TID 160928. ΠΟ 170000 and ΠΟ 225218). a composite spectrum cannot be avoided with observations from) ground. iustrmucuts uuless it can be demonstrated that the companion bIuniuositv is wach weaker.," The contamination on the observed spectra depends on the luminosity difference of the components and on the slit width of the spectrograph on the sky; for several stars the observed spectra are expected to be affected by the companion star and, in some cases (HD 290492, HD 38545, HD 47152, HD 141851, HD 153808, HD 159082, HD 160928, HD 170000 and HD 225218), a composite spectrum cannot be avoided with observations from ground instruments unless it can be demonstrated that the companion luminosity is much weaker."430 The effect of the secondary star should uuderlie the discordant classifications proposed for ΠΟ 225215. either BO Π or À3 Vs: moreover its UV uaenitudes. as determined by the TDL experiment (Thompson ot al.," The effect of the secondary star should underlie the discordant classifications proposed for HD 225218, either B9 III or A3 Vs; moreover its UV magnitudes, as determined by the TD1 experiment (Thompson et al."431 1978). would sugeest a immuch lower reddening than that derived roni the Stróuuugreu photoietry. ie. E(b-v)-0.03. a value more coherent with the stellar visual magnitude.," 1978), would suggest a much lower reddening than that derived from the Strömmgren photometry, i.e. E(b-y)=0.03, a value more coherent with the stellar visual magnitude."432 The two stars ΠΟ 111851 and WD 119303. beiuz X-ray sources (ITIüuusch et al.," The two stars HD 141851 and HD 149303, being X-ray sources (Hünnsch et al."433 1998). are expected o have cool conipaauiots.," 1998), are expected to have cool companions."434 Significant discrepancies in the magnitude anc photometric colours of the AB system ΠΟ 193281. are reported in the literature (BSC)., Significant discrepancies in the magnitude and photometric colours of the AB system HD 193281 are reported in the literature (BSC).435 The ESA llipparcos Catalogue (1997) allowed to colmplete information on sole stars aud to add new binaries: these data are collected iu cobuuus 8 aud 9 of Table 1., The ESA Hipparcos Catalogue (1997) allowed to complete information on some stars and to add new binaries; these data are collected in columns 8 and 9 of Table 1.436 Further known binaries are: and UD 81918 and ΠΟ 171918., Further known binaries are: $\bullet$ The already quoted spectroscopic binary HD 111786 and HD 84948 and HD 171948.437 The 1 components of the last two binaries are all stars. according to Pauuzeu ct al (1998): however. we note that the Me does not slow uuderabundance higher than the other metals aud that the abuudauces of the key elements C.N.O and S are not evel.," The 4 components of the last two binaries are all stars, according to Paunzen et al (1998); however, we note that the Mg does not show underabundance higher than the other metals and that the abundances of the key elements C,N,O and S are not given."438 e IID 112703 is a suspected occultation double (BSC)., $\bullet$ HD 142703 is a suspected occultation double (BSC).439 ο IID 79108 is a suspected SB (BSC). but we could uot retrive any further information on the possible iuflueuce of the companion on photometric aud spectroscopic data.," $\bullet$ HD 79108 is a suspected SB (BSC), but we could not retrive any further information on the possible influence of the companion on photometric and spectroscopic data."440 e The oxveen spectruu indicates that ΠΟ 119303 is au SB system (Paunzen ct al 1999)., $\bullet$ The oxygen spectrum indicates that HD 149303 is an SB system (Paunzen et al 1999).441 We also note the incousistent classifications assigued to TD 22170 1.0. a star CAM) or Te-weak with variable iuteusitv of the SUT 1125-30 doublet (Grav 1988)., We also note the inconsistent classifications assigned to HD 22470 i.e. a star (AM) or He-weak with variable intensity of the SiII 4128-30 doublet (Gray 1988).442 The explanation of the peculiarities has been given by the Hipparcos detection of its duplicity (sce colts 8 and 9 of Table 1)., The explanation of the peculiarities has been given by the Hipparcos detection of its duplicity (see columns 8 and 9 of Table 1).443 Similar remarks apply to two other binaries. TD 17152 and IID 170000. which have been both classified. as single objects. either Ap or.," Similar remarks apply to two other binaries, HD 47152 and HD 170000, which have been both classified, as single objects, either Ap or."444. Two more spurious candidates are TD 130155 which is in reality an Ap Si-A L200 star (Caay 1988) aud IID 159082 which is a BO ITe-Mu (see. for example. EHülbrie Mathys 1996).," Two more spurious candidates are HD 130158 which is in reality an Ap $\lambda$ 4200 star (Gray 1988) and HD 159082 which is a B9 Hg-Mn (see, for example, Hubrig Mathys 1996)."445 The peculiarity of ΠΟ 108282. which has a very high cy value aud therefore a derived ee which is the lowest of our sample stars. but also a very ligh v sini value. deserves further analysis before being assigned to the class: Tanck et al. (," The peculiarity of HD 108283, which has a very high $_1$ value and therefore a derived g which is the lowest of our sample stars, but also a very high v sini value, deserves further analysis before being assigned to the class; Hauck et al. ("4461998) rejected it from the class.,1998) rejected it from the class.447 We couchide that from erouud aud space observations close duplicity. which is able to affect the observed spectrin has been already observed or suspected for 2 1% or 33 Á( (if the stars classified U by Hipparcos are included) of the stars of Table 1 aud that some spurious candidates are present.," We conclude that from ground and space observations close duplicity which is able to affect the observed spectrum, has been already observed or suspected for 24 $\%$ or 33 $\%$ (if the stars classified U by Hipparcos are included) of the stars of Table 1 and that some spurious candidates are present."448 Tu the abundance analyses À Boo stars have been generally considered as sinele stars., In the abundance analyses $\lambda$ Boo stars have been generally considered as single stars.449 The exception is the analysis of the two SB2 stars ΠΟ 819[8 aud IID 1719£8 by Pauuzeu et al (1998)., The exception is the analysis of the two SB2 stars HD 84948 and HD 171948 by Paunzen et al (1998).450 In the present section we consider the influence of a possible companion on model parameters aud on derived abundances and discuss the ability to pick out the alveady known binarics ou the basis of spectra., In the present section we consider the influence of a possible companion on model parameters and on derived abundances and discuss the ability to pick out the already known binaries on the basis of spectra.451 This has a direct. bearing on the issue whether binaritv may lead to significant systematic errors in the abundance analysis and classification of A Boo stars., This has a direct bearing on the issue whether binarity may lead to significant systematic errors in the abundance analysis and classification of $\lambda$ Boo stars.452discussion later).,discussion later).453 In view of all this. we shall consider Ha emission line equivalent width variations of more than as intrinsic. as can be seen in the figures.," In view of all this, we shall consider $\alpha$ emission line equivalent width variations of more than as intrinsic, as can be seen in the figures."454 The smallness of the contamination is understandable. following the expansion and fading of the nebula. as even our earliest observations were already about 35 years after the outburst of the nova.," The smallness of the contamination is understandable, following the expansion and fading of the nebula, as even our earliest observations were already about 35 years after the outburst of the nova."455 As the line wings are always faint. the value of the total He equivalent width is dominated by the value of the narrow central component. so the vartations of these two components are similar.," As the line wings are always faint, the value of the total $\alpha$ equivalent width is dominated by the value of the narrow central component, so the variations of these two components are similar."456 However it is not clear to what extent any of the orbital phase variations are significant., However it is not clear to what extent any of the orbital phase variations are significant.457 In some cases. like November 2003 (Fig. 7))," In some cases, like November 2003 (Fig. \ref{nov03}) )"458 or October 2005. the variations are small: for instance. on November 1-2 2003. the ratio between the largest and smallest value of equivalent width is only 1.13.," or October 2005, the variations are small: for instance, on November 1-2 2003, the ratio between the largest and smallest value of equivalent width is only 1.13."459 This can give a limit to deviations from circular symmetry at that time., This can give a limit to deviations from circular symmetry at that time.460 In other cases. e.g. September 2005 (Fig. 8)).," In other cases, e.g. September 2005 (Fig. \ref{earlysep05}) ),"461 or August 2008. the variations. and the scatter are much larger.," or August 2008, the variations, and the scatter are much larger."462 As in those longer series. more observing points are available. these larger variations are probably indicative of rapid variations from night to night.," As in those longer series, more observing points are available, these larger variations are probably indicative of rapid variations from night to night."463 There are however significant differences in average equivalent widthes from one epoch to another. as is illustrated in Fig.," There are however significant differences in average equivalent widthes from one epoch to another, as is illustrated in Fig."464 9 for the central component: lower values are found for the 2003 epoch. with comparatively stable values over different phases. while higher values are found in September 2005. and a much larger change with phase in 2008.," \ref{TotalFour} for the central component: lower values are found for the 2003 epoch, with comparatively stable values over different phases, while higher values are found in September 2005, and a much larger change with phase in 2008."465 Quite large. longer time scale variations of the total and line centre equivalent widths are seen (e.g. Fig.," Quite large, longer time scale variations of the total and line centre equivalent widths are seen (e.g. Fig."466 13. for the central component)., \ref{centreall} for the central component).467 The values are clearly smaller in 2003 and 2004 than earlier or later., The values are clearly smaller in 2003 and 2004 than earlier or later.468 Such variations would be hard to explain by contamination from varying contributions of flux from the ejected nebula., Such variations would be hard to explain by contamination from varying contributions of flux from the ejected nebula.469 For the wings. no clear systematic. variation with time of their equivalent width can be established. as the scatter is large: but changes in their profiles and/or terminal velocity are definitely seen.," For the wings, no clear systematic variation with time of their equivalent width can be established, as the scatter is large; but changes in their profiles and/or terminal velocity are definitely seen."470 The central component of the Ha profile is assymmetric and varies with orbital phase. suggesting something like the well known S wave of cataclysmic variables.," The central component of the $\alpha$ profile is assymmetric and varies with orbital phase, suggesting something like the well known S wave of cataclysmic variables."471 Comparison of profiles from successive spectra. obtained at some observational epochs when fairly long time seres of observations could be made. reveals this effect.," Comparison of profiles from successive spectra, obtained at some observational epochs when fairly long time series of observations could be made, reveals this effect."472 Three examples are displayed. from the beginning. middle and end of the campaign. in Fig. 10..," Three examples are displayed, from the beginning, middle and end of the campaign, in Fig. \ref{SwaveB},"473 11 and 12. respectively. where series of spectra from the same epoch have been bined over 0.1 intervals of phases.," \ref{SwaveC} and \ref{Swave} respectively, where series of spectra from the same epoch have been bined over 0.1 intervals of phases."474 They show the permanency. and the stability of the phenomenon.," They show the permanency, and the stability of the phenomenon."475" The detailed profiles may not look exactly the same. for a given bin. at each of the three epochs displayed. but these small differences are due to the averaging process: although the total number of observed spectra was quite large. the number of spectra available for a given bin at a given epoch ts small. varying between | and 5. and furthermore the phase distribution within a given bin is quite inhomogeneous. so that the ""average"" aspect can be strongly influenced by the quality or phase of a single spectrum in the average."," The detailed profiles may not look exactly the same, for a given bin, at each of the three epochs displayed, but these small differences are due to the averaging process: although the total number of observed spectra was quite large, the number of spectra available for a given bin at a given epoch is small, varying between 1 and 5, and furthermore the phase distribution within a given bin is quite inhomogeneous, so that the ""average"" aspect can be strongly influenced by the quality or phase of a single spectrum in the average."476 Nevertheless it is clear that the main features. like the phase of the maximum blue or red peak in the central component ts quite clear. and reproducible.," Nevertheless it is clear that the main features, like the phase of the maximum blue or red peak in the central component is quite clear, and reproducible."477 This confirms also that the spectroscopically determined period of Kürrster and Barwig (1988)) is the most appropriate for our analysis., This confirms also that the spectroscopically determined period of Kürrster and Barwig \cite{Kurs88}) ) is the most appropriate for our analysis.478 The three spectra shown in Fig., The three spectra shown in Fig.479 4. are rather similar: the lower one was obtained at the OHP on Dec. 10. 2007 through thin clouds (exposure of IOmn). while the upper one was obtained at the OHP on Dec. 7. 2008 in good conditions," \ref{low} are rather similar: the lower one was obtained at the OHP on Dec. 10, 2007 through thin clouds (exposure of 10mn), while the upper one was obtained at the OHP on Dec. 7, 2008 in good conditions"480When (he chemical evolution model is constrained by observations of (he diffuse gas. one can predict the [C-O/II] distribution in stars. assuming that Iuminous long-lived remnants were produced.,"When the chemical evolution model is constrained by observations of the diffuse gas, one can predict the [C-O/H] distribution in stars, assuming that luminous long-lived remnants were produced."481 Further. the stars should also show appropriate [Si/C] signatures.," Further, the stars should also show appropriate [Si/C] signatures."482 If such stars were formed. we can reasonably expect to find them among (he various samples of extremely metal poor stars which have been spectroscopically analyzed.," If such stars were formed, we can reasonably expect to find them among the various samples of extremely metal poor stars which have been spectroscopically analyzed."483 The data discussed below follows from earlier work by Gratton Sueden (1983. 1991). MeWillium et ((1995). Divan. Norris Beers (1996). and others.," The data discussed below follows from earlier work by Gratton Sneden (1988, 1991), McWilliam et (1995), Ryan, Norris Beers (1996), and others."484 llere we are concerned primarily with the abundances of C. O. Mg. Si. and Ee.," Here we are concerned primarily with the abundances of C, O, Mg, Si, and Fe."485 Unfortunately. not all of these abundances are available from every study. and care must be exercised when interpreting and combining these results. since the samples are subject to strong observational selection effects.," Unfortunately, not all of these abundances are available from every study, and care must be exercised when interpreting and combining these results, since the samples are subject to strong observational selection effects."486" In. addition. most of the work has been done on giant stars whose atmospheres are subject to non-LTE effects. possible mixing, effects. 1-D versus 3-D effects. and incompleteness due to the weakness of lines analyzed."," In addition, most of the work has been done on giant stars whose atmospheres are subject to non-LTE effects, possible mixing effects, 1-D versus 3-D effects, and incompleteness due to the weakness of lines analyzed."487 Carbon amd oxygen are ραασαν vulnerable., Carbon and oxygen are particularly vulnerable.488 Our main sources of data are (he work of Cavrel et ((2004) and a further discussion of the carbon and oxveen abundances of the same stars by Spite et ((2005)., Our main sources of data are the work of Cayrel et (2004) and a further discussion of the carbon and oxygen abundances of the same stars by Spite et (2005).489" For carbon [C'/Fe]. the sources are the unmixed stars of Spite et ((2005). Akerman οἱ ((2004). as well as those stars from Honda et ((2004) and Barklem οἱ ((2005) with log L/L.S2.3 (caleulated from log T,y. log g. and assumed mass of 0.8 M. )."," For carbon [C/Fe], the sources are the unmixed stars of Spite et (2005), Akerman et (2004), as well as those stars from Honda et (2004) and Barklem et (2005) with log $_{\odot}\lesssim2.3$ (calculated from log $_{eff}$, log g, and assumed mass of 0.8 $_{\odot}$ )."490 This selection criterion was used in order (ο reduce (he possibiltv of including mixed-CNO atmospleres., This selection criterion was used in order to reduce the possibilty of including mixed-CNO atmospheres.491 Note that all sources. except Darklem et al.," Note that all sources, except Barklem et al.,"492 avoided including carbon-rich stars., avoided including carbon-rich stars.493 Data for oxvgen [(O/Fe]| are from Akerman οἱ ((2004) and Spite et ((2005)., Data for oxygen [O/Fe] are from Akerman et (2004) and Spite et (2005).494 Following Spite et al.," Following Spite et al.,"495 their [O/Fe] data were lowered by 0.25 dex to agree with the Akerman et delata., their [O/Fe] data were lowered by 0.25 dex to agree with the Akerman et data.496 Magnesium abundances [Mg/Fe| are available from all sources except Akerman et al..," Magnesium abundances [Mg/Fe] are available from all sources except Akerman et al.,"497 silicon Si/Fe] from all sources except Akerman et aand. Barklem et al..," silicon [Si/Fe] from all sources except Akerman et and Barklem et al.,"498 while values of |Fe/1l) are available for all stars., while values of [Fe/H] are available for all stars.499 Generally. (he above samples contain some of the most metal poor stars known.," Generally, the above samples contain some of the most metal poor stars known."500 Most of these stars are more metal poor than the metal poorest globular clusters (|Fe/II]o—2.4. (O/I]e —1.8).," Most of these stars are more metal poor than the metal poorest globular clusters $\sim-2.4$, $\sim-1.8$ )."501 As such they are assumed to have comparable or older ages (~13—14x10? vis) and hence masses  0.3M..., As such they are assumed to have comparable or older ages $\sim13-14 \times10^{9}$ yrs) and hence masses $\sim0.8$ $_{\odot}$.502 The data are presented as plots of [O/II] and [C/I] versus ο). |S1/IH]. and |Mg/1l| in I13.," The data are presented as plots of [O/H] and [C/H] versus [Fe/H], [Si/H], and [Mg/H] in 1–3."503 The solid line in each figure shows the Population II sequence., The solid line in each figure shows the Population II sequence.504 It is defined bv the apparent clumping of many of the stars (especially at the metal rich end) and from the figures. by the following implied abundance ratios (ie. |C/Fe]|e0.25. [O/Fe|-0.65. \e/Fel~0.30. and [Si/Fe]~ 0.35).," It is defined by the apparent clumping of many of the stars (especially at the metal rich end) and from the figures, by the following implied abundance ratios (i.e. $\sim0.25$, $\sim0.65$, $\sim0.30$, and $\sim0.35$ )."505 These ratios are similar to those found in earlier studies ol the less metal poor Pop L stars (e.g. the works cited in (he first paragraph of Chis section)., These ratios are similar to those found in earlier studies of the less metal poor Pop II stars (e.g. the works cited in the first paragraph of this section).506would be 2300 highredshift. 22.2. quasars with over 500 detectable DLAs.,"would be $\sim 2300$ high–redshift, $z>2.2$, quasars with over 500 detectable DLAs."507 If there is a serious bias in existing optical samples of DLAs the number of DLAs detected will be larger., If there is a serious bias in existing optical samples of DLAs the number of DLAs detected will be larger.508 Since W=16.0 corresponds to R=1s8.2 for a tvpical quasar. quasars that are substantially reddened. &2 mag in the optical. would still be bright enough for highresolution spectroscopic follow.up.," Since K=16.0 corresponds to R=18.2 for a typical quasar, quasars that are substantially reddened, $\la5092\,$ mag in the optical, would still be bright enough for high–resolution spectroscopic follow–up."510 An alternative strategy. could be to use radioselected quasars to undertake a survey for DLAs unbiased by extinction due to dust., An alternative strategy could be to use radio–selected quasars to undertake a survey for DLAs unbiased by extinction due to dust.511 Phere are two disadvantages to this approach., There are two disadvantages to this approach.512 Firstly there are insullicient highredshift racio quasars to vield a sample of DLAs of the size envisaged here., Firstly there are insufficient high–redshift radio quasars to yield a sample of DLAs of the size envisaged here.513 Secondly. a significant fraction. of the quasars will be very faint in the optical. precluding highresolution spectroscopy. vet it ds essential to survey the spectra. of all the quasars for the sample to be unbiased.," Secondly, a significant fraction of the quasars will be very faint in the optical, precluding high–resolution spectroscopy, yet it is essential to survey the spectra of all the quasars for the sample to be unbiased."514 Nevertheless. any subset of quasars sulfering very large extinetions that still elude the Ix.band lux.limited selection could be identified in a radio SULVON.," Nevertheless, any subset of quasars suffering very large extinctions that still elude the K–band flux–limited selection could be identified in a radio survey."515 Exclucling lensing by ealaxy clusters. there are currently only some 40 examples of strong gravitational lensing known 4).," Excluding lensing by galaxy clusters, there are currently only some 40 examples of strong gravitational lensing known )."516 For the same survey for bright quasars considered above. Ay<16.0. counting quasars of all redshifts there will be over 10000 quasars over the 4000 deg? survey area.," For the same survey for bright quasars considered above, $K<16.0$, counting quasars of all redshifts there will be over 10000 quasars over the 4000 $^2$ survey area."517 For a typical quasar colour. V-]x22.5 the survey magnitude limit is equivalent to V—18.5., For a typical quasar colour V-K=2.5 the survey magnitude limit is equivalent to V=18.5.518 Therefore the apparent magnitudes of the quasars will be similar to the apparent magnitudes of the quasars observed in the LIST. snapshot survey (Maoz et al 1993) which have |—ISOHEOLS., Therefore the apparent magnitudes of the quasars will be similar to the apparent magnitudes of the quasars observed in the HST snapshot survey (Maoz et al 1993) which have $V=18.0\pm0.8$.519 There are five cases of gravitational lensing amongst the 502 quasars imaged in the snapshot survey., There are five cases of gravitational lensing amongst the 502 quasars imaged in the snapshot survey.520 Because the apparent. magnitudes are similar we can use the results of the snapshot survey to estimate the number of lenses in the ΝΗΣΙ survey. by assuming that 1% of the quasars are lensecl (the fraction of lenses in a sample depends on the sample depth because of magnification bias)., Because the apparent magnitudes are similar we can use the results of the snapshot survey to estimate the number of lenses in the UKIRT survey by assuming that $1\%$ of the quasars are lensed (the fraction of lenses in a sample depends on the sample depth because of magnification bias).521 Therefore the UINIICE. survey. should. produce roughly 100 eravitational lenses., Therefore the UKIRT survey should produce roughly 100 gravitational lenses.522 The number of lenses detected could be larger than LOO if bias due to dust is important., The number of lenses detected could be larger than 100 if bias due to dust is important.523 ]t is unlikely that. the. image quality achieved. in the survey will be good enough to detect examples. of eravitational lensing where the image separation is less than (0.5 aresec.," It is unlikely that the image quality achieved in the survey will be good enough to detect examples of gravitational lensing where the image separation is less than $0.5\,$ arcsec."524 Sensitivity to separations as small as 0.2 arcsec could be achieved by later imaging all the bright quasars using adaptive optics.," Sensitivity to separations as small as $0.2\,$ arcsec could be achieved by later imaging all the bright quasars using adaptive optics."525 The fraction of lenses in a IEx-selected sample could be increased. in two wavs., The fraction of lenses in a K-selected sample could be increased in two ways.526 Firstly. a sample of similar size could be obtained by surveying a larger area of sky but to a brighter magnitude limit. thereby increasing the magnification bias.," Firstly, a sample of similar size could be obtained by surveying a larger area of sky but to a brighter magnitude limit, thereby increasing the magnification bias."527 Alternatively. followup highresolution imaging could be limited to quasars of high redshift. since the probability that a quasar is lensed increases with redshift.," Alternatively, follow–up high–resolution imaging could be limited to quasars of high redshift, since the probability that a quasar is lensed increases with redshift."528 The interpretation of a survey for gravitational lenses requires knowledge of the quasar luminosity function ab magnitudes fainter than the search limit. since the lensecl quasars have been magnified.," The interpretation of a survey for gravitational lenses requires knowledge of the quasar luminosity function at magnitudes fainter than the search limit, since the lensed quasars have been magnified."529 A potential drawback of the WN method is contamination of the sample of candidate quasars at faint magnitudes by compact galaxics that morphologically cannot be distinguished. from. stars., A potential drawback of the KX method is contamination of the sample of candidate quasars at faint magnitudes by compact galaxies that morphologically cannot be distinguished from stars.530 Nevertheless. at these fainter magnitudes where the surface densities are higher it will be feasible to measure the quasar Luminosity function. based on spectroscopic surveys of complete Huxlimited. samples (ic. with no colour or morphological selection) using multiobject spectrographs.," Nevertheless, at these fainter magnitudes where the surface densities are higher it will be feasible to measure the quasar luminosity function based on spectroscopic surveys of complete flux–limited samples (i.e. with no colour or morphological selection) using multi–object spectrographs."531 One advantage of Ix.band: survevs for eravitational lenses over racio surveys such as CLASS (Jackson ct al 1998) is the case with which the redshift distribution of the unlensed source population can be measured., One advantage of K–band surveys for gravitational lenses over radio surveys such as CLASS (Jackson et al 1998) is the ease with which the redshift distribution of the unlensed source population can be measured.532 We thank Scott Croom lor comments on the clralt., We thank Scott Croom for comments on the draft.533 The authors acknowledge the data and analysis facilities provided by the Starlink Project which is run by CCLRC on behalf of PPARC., The authors acknowledge the data and analysis facilities provided by the Starlink Project which is run by CCLRC on behalf of PPARC.534"with stellar parameters Tig=4600 K, logg=1.6 [cgs], and scaled solar chemical composition with ","with stellar parameters $T_\mathrm{eff}=4600$ K, $\log{g}=1.6$ [cgs], and scaled solar chemical composition with $=-3$."535[Fe/H]——3. fram .Barhiyetal. heisurface gravity based odel, The stellar parameters were taken from \cite{barbuy03} with a slight revision of the surface gravity based on the parallax determination of HD122563 from the data reduction by \cite{vanleeuwen07}.536 wag’ computed using a] discrete 480x480 x240grid representing ia cube of physical dimensiéii8197003700x1100 Mme., The 3D model was computed using a discrete $480{\times}480{\times}240$ grid representing a cube of physical dimensions $3700{\times}3700{\times}1100$ $^3$.537" The equations of mass, energy, and momentum wete solved together witiethe radiative transfer consdirationequations dopting isti i"," The equations of mass, energy, and momentum conservation were solved together with the radiative transfer equations adopting realistic equation of state and opacities."538 iti imulation -s8g0ence 606ed fanothe-poesent work dqüGists of 51 snapshots that hours of stellar time and it typically contains about 10 Vetweital{outs150granules at the surface., The simulation sequence used for the present work consists of 51 snapshots that cover about 150 hours of stellar time and it typically contains about 10 granules at the surface.539 The evolution timescale of the granules is about 90 hours., The evolution timescale of the granules is about 90 hours.540" For more details, see Colletetal.(2009)."," For more details, see \cite{collet09}."541. We used the3D model to synthesize all 58 lines shown in Fig 1.., We used the3D model to synthesize all 58 lines shown in Fig \ref{f:bisectors}. .542 For the line synthesis we adopted a coarser grid of 5050x125 points to which we interpolated the original simulation., For the line synthesis we adopted a coarser grid of $50{\times}50{\times}125$ points to which we interpolated the original simulation.543 A number of tests were performed with a higher-resolution grid of 100x100x125 points to ensure that this had no significant impact on the shapes of the synthetic line profiles., A number of tests were performed with a higher-resolution grid of $100{\times}100{\times}125$ points to ensure that this had no significant impact on the shapes of the synthetic line profiles.544" The synthetic 3D profiles were convolved with a Gaussian instrumental profile of FWHM=1.5 kmss71, which corresponds to the spectral resolution of our data."," The synthetic 3D profiles were convolved with a Gaussian instrumental profile of $\mathrm{FWHM}=1.5$ $^{-1}$, which corresponds to the spectral resolution of our data."545" From x? line profile fitting of about 30 llines awith the highest local with Vsini and iron abundance as free parameters, S/N,we obtained Vsini=3.2+0.6 ss-!."," From a $\chi^2$ line profile fitting of about 30 lines with the highest local $S/N$ , with $V\sin i$ and iron abundance as free parameters, we obtained $V\sin i=3.2\pm0.6$ $^{-1}$."546" For the x? fitting we convolved the disk- profiles (e.g.,Gray1992,p."," For the $\chi^2$ fitting we convolved the disk-integrated profiles \citep[e.g.,][p.\,370]{gray92:book}."547 To calculate the disk-integrated profiles for comparison370).. with the observational data we applied a Vsini=3.2 ss! to the emergent intensities., To calculate the disk-integrated profiles for comparison with the observational data we applied a $V\sin i=3.2$ $^{-1}$ to the emergent intensities.548 A detailed abundance analysis of our 1122563 spectrum using the 3D model will be provided in a later In Fig., A detailed abundance analysis of our 122563 spectrum using the 3D model will be provided in a later In Fig.549 2 we show a comparison of observations and model for 3 representative llines., \ref{f:bis} we show a comparison of observations and model for 3 representative lines.550" The observed lines have been shifted by about +30, +30, and ((from strongest to weakest line) to make their core velocities agree with the theoretical values."," The observed lines have been shifted by about +30, +30, and (from strongest to weakest line) to make their core velocities agree with the theoretical values."551 This allows a better comparison of line bisector because it removes the uncertainties in the measured core wavelengths and rest laboratory wavelengths., This allows a better comparison of line bisector because it removes the uncertainties in the measured core wavelengths and rest laboratory wavelengths.552" We discuss the core wavelength shifts below and show that the artificial shifts applied here are within the observational The average difference in bisector velocities (model minus observations for all 58 llines) is 10+71 ((hereafter, error bars correspond to the lo scatter)."," We discuss the core wavelength shifts below and show that the artificial shifts applied here are within the observational The average difference in bisector velocities (model minus observations for all 58 lines) is $10\pm71$ (hereafter, error bars correspond to the $\sigma$ scatter)."553" 'This number, however, depends on the relative flux."," This number, however, depends on the relative flux."554" Near the continuum, the bisector point-by-point difference is about —30+145 wwhile it is 9-33 aat a residual flux of about 0.2."," Near the continuum, the bisector point-by-point difference is about $-30\pm145$ while it is $9\pm33$ at a residual flux of about 0.2."555 At all flux values the difference is consistent with zero within the lo scatter.," At all flux values the difference is consistent with zero within the $1\,\sigma$ scatter."556" The formal error in the observed bisector velocities shows a residual flux dependency remarkably similar to that described above; i.e., it ranges from about 30 to 150 ffrom core to continuum."," The formal error in the observed bisector velocities shows a residual flux dependency remarkably similar to that described above; i.e., it ranges from about 30 to 150 from core to continuum."557" Thus, the scatter seen in the differences between theoretical and observed bisectors is fully compatible with the 1o observational errors."," Thus, the scatter seen in the differences between theoretical and observed bisectors is fully compatible with the $1\,\sigma$ observational errors."558" Measurements of core wavelength shifts (Δυο) are in principle more precise than line bisectors, since only a small portion of the line profile is examined, thus lowering the probability of being affected by blends."," Measurements of core wavelength shifts $\Delta v_c$ ) are in principle more precise than line bisectors, since only a small portion of the line profile is examined, thus lowering the probability of being affected by blends."559 In Fig., In Fig.560" 3 we show the relation between Av, and line strength.", \ref{f:shifts} we show the relation between $\Delta v_c$ and line strength.561" The so-called “third signature"" of stellar granulation (Gray 2009),, i.e., the increased blueshift with decreasing line strength, is clearly detected in the data and very closely reproduced by the 3D model."," The so-called “third signature” of stellar granulation \citep{gray09}, i.e., the increased blueshift with decreasing line strength, is clearly detected in the data and very closely reproduced by the 3D model."562" The third signature is due to the fact that weaker lines form in deeper layers, where the granulation velocities and intensity contrast are larger."," The third signature is due to the fact that weaker lines form in deeper layers, where the granulation velocities and intensity contrast are larger."563 The line-to-line scatter seen in Fig., The line-to-line scatter seen in Fig.564" 3 increases towards weaker lines, an effect that is present also in the 3D model predictions."," \ref{f:shifts} increases towards weaker lines, an effect that is present also in the 3D model predictions."565" It is tempting to conclude that this increased scatter is real, but given the timescale of the evolution of granules in comparison with the duration of the simulation and the possibility that line cores of weaker lines are more difficult to measure (but see we must refrain from making a strong statement about below),this finding."," It is tempting to conclude that this increased scatter is real, but given the timescale of the evolution of granules in comparison with the duration of the simulation and the possibility that line cores of weaker lines are more difficult to measure (but see below), we must refrain from making a strong statement about this finding."566" Note also that the variation of the continuous opacity over the wavelengthrange covered by our data is substantial, which introduces wavelength dependent shifts that might contribute to the observed scatter (e.g.,Dravinsetal. 1981)."," Note also that the variation of the continuous opacity over the wavelengthrange covered by our data is substantial, which introduces wavelength dependent shifts that might contribute to the observed scatter \citep[e.g.,][]{dravins81}. ."567". The difference between observations and model predictions forAv, has a lo scatter of ms..", The difference between observations and model predictions for$\Delta v_c$ has a $\sigma$ scatter of .568Q breaks when the potential surpasses the spontaneous pair formation threshold.,$\bar Q$ breaks when the potential surpasses the spontaneous pair formation threshold.569 As a result. (L(7)) is very small but finite for T'<Ty.," As a result, $\langle L(T) \rangle$ is very small but finite for $T < T_L$."570 The breaking of chiral svmunetry is indicated by a finite value of the chiral condensate WO)=(ee)~Mj. which measures the dvnaniucally generated (constituent) quark mass Af. obtained for a Lagrangian with massless quarks.," The breaking of chiral symmetry is indicated by a finite value of the chiral condensate $\chi(T) \equiv \langle \bar{\psi} \psi \rangle \sim M_q$, which measures the dynamically generated (“constituent”) quark mass $M_q$, obtained for a Lagrangian with massless quarks."571 At high temperature. this mass melts. so that VG) delines the chiral svnunetry restoration temperature TV.," At high temperature, this mass melts, so that (T) defines the chiral symmetry restoration temperature $T_{\chi}$."572 Here we have exact chiral svaumetry onlv if the input quarks are massless: for finite quark mass. (he svimmnmieiry remains explicitly broken. so that then 4(7) only becomes very small for T>TV.," Here we have exact chiral symmetry only if the input quarks are massless; for finite quark mass, the symmetry remains explicitly broken, so that then $\chi(T)$ only becomes very small for $T > T_{\chi}$."573 Both (L(T)) and 4(7) have been studied extensively in finite temperature lattice QCD al vanishing overall barvon number., Both $\langle L(T) \rangle$ and $\chi(T)$ have been studied extensively in finite temperature lattice QCD at vanishing overall baryon number.574" The corresponding susceptibilities (derivatives wilh respect lo T) peak sharply. defining Tj, and 7i. and within errors. the (wo temperatures and hence the (wo phenomena (deconfinement and chiral syamietry restoration) coincide."," The corresponding susceptibilities (derivatives with respect to $T$ ) peak sharply, defining $T_L$ and $T_{\chi}$, and within errors, the two temperatures and hence the two phenomena (deconfinement and chiral symmetry restoration) coincide."575 The critical temperature for the resulting transition from hadronic matter to QGP. for two light quark flavors. is thus determined as T.7175 MeV. Tt was clear from the very first finite temperature lattice studies (hat the deconfined medium in the region above T; is verv strongly interacting and (hus quite far [from an ideal plasma. [1]..," The critical temperature for the resulting transition from hadronic matter to QGP, for two light quark flavors, is thus determined as $T_c \simeq 175$ MeV. It was clear from the very first finite temperature lattice studies that the deconfined medium in the region above $T_c$ is very strongly interacting and thus quite far from an ideal plasma \cite{inter-plasma}."576 ‘This is best seen from the interaction measure (the trace of tensor). defined as A=!2o.," This is best seen from the interaction measure (the trace of energy-momentum tensor), defined as =."577 For non-interactinge massless constituents (the “conformal” limit). A=0. so that the temperature is (he only scale.," For non-interacting massless constituents (the “conformal” limit), $\Delta \equiv 0$, so that the temperature is the only scale."578 The quarks ancl gluons of QCD are ideally massless. but the so-called (race anomaly violates conformality and introduces a dimensional scale.," The quarks and gluons of QCD are ideally massless, but the so-called trace anomaly violates conformality and introduces a dimensional scale."579 The behavior of A(Z) is shown in reldella both in pure gauge theory for different color groups and for fll QCD with different [Iavor content.," The behavior of $\Delta(T)$ is shown in \\ref{delta}580 both in pure gauge theory for different color groups and for full QCD with different flavor content."581 For sufficiently high temperature. asymptotic freedom is expected to result in an ideal QGP.," For sufficiently high temperature, asymptotic freedom is expected to result in an ideal QGP."582 How hieh does Z' have to be in order to allow some form of a weak coupling expansion (perturbation theory) to describe the approach to this limit?, How high does $T$ have to be in order to allow some form of a weak coupling expansion (perturbation theory) to describe the approach to this limit?583"for the orbital period of the light curve (the photoelectric data, although covering a larger total timespan, are too scarcely sampled for that purpose).","for the orbital period of the light curve (the photoelectric data, although covering a larger total timespan, are too scarcely sampled for that purpose)."584" The analysis was made preferentially in the V band, where more data are available, but the R band data were used also to check for possible wavelength-dependant effects."," The analysis was made preferentially in the V band, where more data are available, but the R band data were used also to check for possible wavelength-dependant effects."585" They were analysed using the ""Period04"" PC code from Lenz Breger (2004)) and also a programme based on the Deming (1975)) method, in the frequency range 0-50 d-!."," They were analysed using the ""Period04"" PC code from Lenz Breger \cite{Lenz04}) ) and also a programme based on the Deming \cite{Deming75}) ) method, in the frequency range 0-50 $d^{-1}$."586 The resulting periodogramm is displayed in Fig. 5::, The resulting periodogramm is displayed in Fig. \ref{R_Power}:587" the most prominent peak is for a period of 0.214164 days, corresponding, within the uncertainties, to the period obtained by Kürrster and Barwig from their spectroscopic analysis (0.214165), the other peaks being daily aliases."," the most prominent peak is for a period of 0.214164 days, corresponding, within the uncertainties, to the period obtained by Kürrster and Barwig from their spectroscopic analysis (0.214165), the other peaks being daily aliases."588 Our observations folded with this period are shown in Fig. 6.., Our observations folded with this period are shown in Fig. \ref{li-curva}.589 The scatter which remains in this figure is due to the presence of night to night variations and possibly also flickering., The scatter which remains in this figure is due to the presence of night to night variations and possibly also flickering.590" Attempts to fold data with periods corresponding to other peaks in the periodogram lead to unacceptable fits, thus confirming the real period."," Attempts to fold data with periods corresponding to other peaks in the periodogram lead to unacceptable fits, thus confirming the real period."591" If we approximate this light curve with a sine function (thick, red line), the resulting amplitude is A = 0.043 + 0.002."," If we approximate this light curve with a sine function (thick, red line), the resulting semi-amplitude is A = 0.043 $ \pm $ 0.002."592 No other short term variations were found in these observations besides the orbital ones., No other short term variations were found in these observations besides the orbital ones.593" To better detect long timescale variations, such as those found by Bianchini (1987)) in some old novae, our data were corrected for orbital variations."," To better detect long timescale variations, such as those found by Bianchini \cite{Bian87}) ) in some old novae, our data were corrected for orbital variations."594" There is a suspicion of long-term variations with a possible period of around 1500 days (that is about 4 years), already apparent in Fig. 1,,"," There is a suspicion of long-term variations with a possible period of around 1500 days (that is about 4 years), already apparent in Fig. \ref{Overall},"595 but several more years of observations would be needed to confirm this., but several more years of observations would be needed to confirm this.596" The Ha profiles show a well defined central component, with wider wings of lower intensity (Fig. 3)),"," The $\alpha$ profiles show a well defined central component, with wider wings of lower intensity (Fig. \ref{high}) ),"597 suggesting formation in different regions., suggesting formation in different regions.598" In view of this, measurements were made both of the total equivalent width and of that of the line centre only."," In view of this, measurements were made both of the total equivalent width and of that of the line centre only."599" The time of each spectrum (at mid-integration), the corresponding orbital phase according to the ephemeris of Kürrster and Barwig (1988)) and the measured equivalent widths of Ha (both total, and central component only) are given in Table 1."," The time of each spectrum (at mid-integration), the corresponding orbital phase according to the ephemeris of Kürrster and Barwig \cite{Kurs88}) ) and the measured equivalent widths of $\alpha$ (both total, and central component only) are given in Table 1."600 Examples of the variation of the equivalent width of the different Ha line components with Julian date and with Kürrster and Barwig's orbital phase are shown for two epochs in Fig., Examples of the variation of the equivalent width of the different $\alpha$ line components with Julian date and with Kürrster and Barwig's orbital phase are shown for two epochs in Fig.601" 7 and Fig. 8,,"," \ref{nov03} and Fig. \ref{earlysep05},"602 over a time interval not lasting more than a few days., over a time interval not lasting more than a few days.603" Clear differences are seen between these two examples: on one hand, the average value of the Ha equivalent width is much lower at the first epoch; on the other hand, the scatter of the variations with phases is more important during the second epoch."," Clear differences are seen between these two examples: on one hand, the average value of the $\alpha$ equivalent width is much lower at the first epoch; on the other hand, the scatter of the variations with phases is more important during the second epoch."604 Such changes are seen over the whole duration of our observing campaign., Such changes are seen over the whole duration of our observing campaign.605" One concern is a possible, time variable, contamination by emission from the ejected nebula in the Ho line due to the central region (plus a much smaller contribution from the fainter, adjacent emission)."," One concern is a possible, time variable, contamination by emission from the ejected nebula in the $\alpha$ line due to the central region (plus a much smaller contribution from the fainter, adjacent emission)."606" To estimate the possible contribution from the nebula, several tests have been made."," To estimate the possible contribution from the nebula, several tests have been made."607" A detailed map of the nebular emission was obtained by Harman O'Brien (2003)) combining HST images with ground-based, high-resolution spectroscopy."," A detailed map of the nebular emission was obtained by Harman O'Brien \cite{Harm03}) ), combining HST images with ground-based, high-resolution spectroscopy."608" They showed that the nebular emission has an elliptical shape, with major and minor axes of roughly 8.5 x 6.3 arcseconds, with a large Ha contribution coming from a bright rim: these dimensions are much larger than the 3"" entrance aperture, centered on the star, used for our high dispersion spectroscopy, so that the contribution from the nebula should not be significant in the central Ha component."," They showed that the nebular emission has an elliptical shape, with major and minor axes of roughly 8.5 x 6.3 arcseconds, with a large $\alpha$ contribution coming from a bright rim: these dimensions are much larger than the 3"" entrance aperture, centered on the star, used for our high dispersion spectroscopy, so that the contribution from the nebula should not be significant in the central $\alpha$ component."609" The structure of the nebula, as seen in the HST images of Harman O'Brien (2003)), is however clumpy, and several bright knots, located inside the ellipsoid, had been identified."," The structure of the nebula, as seen in the HST images of Harman O'Brien \cite{Harm03}) ), is however clumpy, and several bright knots, located inside the ellipsoid, had been identified."610" To measure their possible contribution to the Ha flux from the central object (integrated over a circular aperture, like the 3"" used), it would be sufficient to measure the intensity of those blobs relative to the one of the central star."," To measure their possible contribution to the $\alpha$ flux from the central object (integrated over a circular aperture, like the 3"" used), it would be sufficient to measure the intensity of those blobs relative to the one of the central star."611" Unfortunately, the central object is saturated in Harman's data, so we cannot estimate this ratio from there."," Unfortunately, the central object is saturated in Harman's data, so we cannot estimate this ratio from there."612" We used instead several ground-based images of ours, obviously of lesser spatial resolution, but where some of the brightest knots can be identified, and found that the intensity ratio between the brightest knots and the central star is typically 1 to 50 in Ha."," We used instead several ground-based images of ours, obviously of lesser spatial resolution, but where some of the brightest knots can be identified, and found that the intensity ratio between the brightest knots and the central star is typically 1 to 50 in $\alpha$."613" It would therefore require several knots within the aperture to make a contribution, and only few are seen in the whole ellipsoid in the HST map."," It would therefore require several knots within the aperture to make a contribution, and only few are seen in the whole ellipsoid in the HST map."614" We used also the spatial resolution along the slit in our low-dispersion spectra to estimate the change in Ha equivalent width when including more and more outer regions, and found also that was a conservative upper-limit for a variable contribution from the nebula (see below the discussion of the low dispersion spectroscopy)."," We used also the spatial resolution along the slit in our low-dispersion spectra to estimate the change in $\alpha$ equivalent width when including more and more outer regions, and found also that was a conservative upper-limit for a variable contribution from the nebula (see below the discussion of the low dispersion spectroscopy)."615" Finally, we note that the measured velocities do not correspond either, at least for the central component, to what is expected from the nebula (see"," Finally, we note that the measured velocities do not correspond either, at least for the central component, to what is expected from the nebula (see"616We thank the stall of the Lord's Bridge observatory for their invaluable assistance in the commissioning and operation of the Arcminute Microkelvin Imager.,We thank the staff of the Lord's Bridge observatory for their invaluable assistance in the commissioning and operation of the Arcminute Microkelvin Imager.617 We thank Nathalie Ysard and John Richer for useful discussions., We thank Nathalie Ysard and John Richer for useful discussions.618 We also thank Simon Casassus. whose comments anc suggestions significantly. improved. this paper.," We also thank Simon Casassus, whose comments and suggestions significantly improved this paper."619 The ΑΔΗ is supported bv Cambridge University and the STEC., The AMI is supported by Cambridge University and the STFC.620 NIIN ancl MED acknowledge the support of PPATIC/STEC studentships., NHW and MLD acknowledge the support of PPARC/STFC studentships.621 and 00(Α where in the second equation we have used the definitions of the angles given in equation (17))., ) and 0 where in the second equation we have used the definitions of the angles given in equation \ref{Angles-Colinear}) ).622 In the degenerate limit when ὁ=z the angular terms become (A19) and," In the degenerate limit when $\phi=\pi$ the angular terms become ^2, ^2 _q^2 and = -3 ^2"623gaps were filled. by [linear interpolation. together with the addition of noise characteristic of the surrounding clata points. in order to eliminate spurious high frequency power in the power spectrum.,"gaps were filled by linear interpolation, together with the addition of noise characteristic of the surrounding data points, in order to eliminate spurious high frequency power in the power spectrum."624 Power spectral densities were derived from. these light. curves using a standard. direct Fourier transform., Power spectral densities were derived from these light curves using a standard direct Fourier transform.625 In Fie., In Fig.626 3. we show the 0.22 and 0 keV PSDs and we see close agreement with the PSDs in the same energy bands., \ref{fig:xmm210psds} we show the 0.2–2 and 2--10 keV PSDs and we see close agreement with the PSDs in the same energy bands.627 As with the PSDs. the low and high energy PPSDs coincide very closely at frequencies below 10 LI. out at higher frequencies the 210 keV. PSD exceeds the ower energv PSD.," As with the PSDs, the low and high energy PSDs coincide very closely at frequencies below $10^{-4}$ Hz, but at higher frequencies the 2–10 keV PSD exceeds the lower energy PSD."628" Note that here. for the low energy baud. we show the 0.22 keV PSD rather than the 0.52 keV PSD and note that the bump centred. around. 10. ""Hz is even ess pronounced in the 0.22 keV band. strengthening our observation that the excess has a strongly energy-dependent shape."," Note that here, for the low energy band, we show the 0.2–2 keV PSD rather than the 0.5–2 keV PSD and note that the bump centred around $10^{-3}$ Hz is even less pronounced in the 0.2–2 keV band, strengthening our observation that the excess has a strongly energy-dependent shape."629 Although not shown. the 00.52 keV. PSD is almost identical to the 0.62 keV. PSD.," Although not shown, the 0.5–2 keV PSD is almost identical to the 0.6–2 keV PSD."630 A simple power law fit to both low and high energy SDs is a very poor fit. leaving alarge residual bump centred. around. 10. νε.," A simple power law fit to both low and high energy PSDs is a very poor fit, leaving alarge residual bump centred around $10^{-3}$ Hz."631 A bending powerlaw. however. which provides a good fit to the high/soft state of (νο N-1 and to the PPSDs of NOGOC4051 (MHardyetal.2004) and 56 (NMLardyetal.2005)... is a &ood fit (Lable 1)).," A bending powerlaw, however, which provides a good fit to the high/soft state of Cyg X-1 and to the PSDs of NGC4051 \citep{mch04} and 6 \citep{mch05}, is a good fit (Table \ref{tab:brpl}) )."632 Tving he break frequencies to be the same in both energy band SDs. we note (as can also be seen clirectly [rom Fig. 3)).," Tying the break frequencies to be the same in both energy band PSDs, we note (as can also be seen directly from Fig. \ref{fig:xmm210psds}) ),"633 hat although the slopes above the break are similar in both energy bands. the slope below the break is noticeably Latter at higher energies.," that although the slopes above the break are similar in both energy bands, the slope below the break is noticeably flatter at higher energies."634 The PPSDs. although less well defined. because of the gap on the orbital time-scales. are quite consistent with the fit parameters given in ‘Table 1..," The PSDs, although less well defined because of the gap on the orbital time-scales, are quite consistent with the fit parameters given in Table \ref{tab:brpl}. ."635 Εις behaviour is cillerent from that of NGCA051 ancl ALCG-6-30-15 (c.g.seeTable6ofAl’Lardyetal.2005). where. although the slopes below the break are not very well determined. there is no evidence for a variation with energy.," This behaviour is different from that of NGC4051 and MCG-6-30-15 \citep[e.g. see Table 6 of][]{mch05} where, although the slopes below the break are not very well determined, there is no evidence for a variation with energy."636 Above the break. however. the slope is Hatter at higher energies in NGC4051 and 30-15.," Above the break, however, the slope is flatter at higher energies in NGC4051 and MCG-6-30-15."637 The PSD slope variations seen here in ccan be more naturally interpreted in terms of a power law (whieh is necessary to explain the power at low freeuencies) together with a Lorentzian component which is stronger at higher energies., The PSD slope variations seen here in can be more naturally interpreted in terms of a power law (which is necessary to explain the power at low frequencies) together with a Lorentzian component which is stronger at higher energies.638 Such a model. in which the power law slope ancl normalisation. and. Lorentzian central frequency are tied. provides an equally good fit to the data.," Such a model, in which the power law slope and normalisation, and Lorentzian central frequency are tied, provides an equally good fit to the data."639 In the case of the latter model. we note that the Lorentzian component is approximately three times stronger in the 2-8.8 keV than 0.52 keV band (Table 2)).," In the case of the latter model, we note that the Lorentzian component is approximately three times stronger in the 2-8.8 keV than 0.5–2 keV band (Table \ref{tab:lor}) )."640 An increasing Lorentzian amplitude at higher energies is also typically observed. in QPOs and some broad Lorentzian components in CLDIIS (e.g.Pottschmidtetal.2003)., An increasing Lorentzian amplitude at higher energies is also typically observed in QPOs and some broad Lorentzian components in GBHs \citep[e.g.][]{pott03}.641. As the exact values of the fit parameters depend on proper fitting of the lower frequency data from aand we leave further discussion to Section 5.., As the exact values of the fit parameters depend on proper fitting of the lower frequency data from and we leave further discussion to Section \ref{s_combinedpsd}.642 There is a residual to the power law plus Lorentzian fit to the PPSD at ~2.2.10! Iz.," There is a residual to the power law plus Lorentzian fit to the PSD at $\sim 2.2 \times64310^{-4}$ Hz."644 This residual can be fitted (using the unbinned data) by the ασος of a second. narrow (width oS10 Pilz). Lorentzian with normalisation approximately half that of the broad. Lorentzian.," This residual can be fitted (using the unbinned data) by the addition of a second, narrow (width $\sim8 \times64510^{-6}$ Hz), Lorentzian with normalisation approximately half that of the broad Lorentzian."646 The fit is shown in Fig. 4.., The fit is shown in Fig. \ref{fig:xmm2lor}. .647 The fit is improved. but. adjusting the F-test probabilities to take account of the fact that theline," The fit is improved but, adjusting the F-test probabilities to take account of the fact that theline"648"They xtvdescribe the inviscid flow of density po, momentum pv, and total energy ej, (including internal, kinetic, and gravitational potential energy).","They describe the inviscid flow of density $\rho$, momentum $\rho \mathbf{v}$, and total energy $e_{\mathit{ikg}}$ (including internal, kinetic, and gravitational potential energy)."649" Further quantities are the velocity vector v, pressure P, and radiative energy flux F’,,."," Further quantities are the velocity vector $\mathbf{v}$, pressure $P$ , and radiative energy flux $F_\mathrm{_{rad}}$."650 The latter is computed from the frequency-integrated intensity J for the gray treatment of opacity and for each wavelength group in the non-gray approach (Sect. ??))., The latter is computed from the frequency-integrated intensity $I$ for the gray treatment of opacity and for each wavelength group in the non-gray approach (Sect. \ref{radsection}) ).651" The gravitational potential is spherical, where Mpor is the mass of the star to be modeled (see Fig. 3., "," The gravitational potential is spherical, where $M_{\mathrm{pot}}$ is the mass of the star to be modeled (see Fig. \ref{1dquantities}, ,"652"bottom right panel); ro and r; are free smoothing parameters: when τό«r*rt, Φ-GM -— while. when r>0, 95-−GM ""E and forr—D~, ®=GM,"," bottom right panel); $r_0$ and $r_1$ are free smoothing parameters: when $r_0^4 \ll r^4 \ll r_1^4$, $\Phi=-\frac{GM_{\mathrm{pot}}}{r}$ ; while when $r\rightarrow 0$, $\Phi\rightarrow \Phi=-\frac{GM_{\mathrm{pot}}}{r_0}$ ; and for $r\rightarrow\infty$, $\Phi\rightarrow \Phi=-\frac{GM_{\mathrm{pot}}}{r_1}$."653" Typically,. το«0.2 Ry, andr;~1.2 R, for RSG simulations T.(?).."," Typically, $r_0\approx0.2$ $R_\star$ and $r_1\approx1.2$ $R_\star$ for RSG simulations \citep{2002AN....323..213F}."654" The numerical simulations described here are performed with OCO?BOLDÓ (OCOnservative COde for the COmputation of COmpressible COnvection in a BOx of L Dimensions, L=2, 3Ó)."," The numerical simulations described here are performed with $^5$ BOLDÓ (ÒCOnservative COde for the COmputation of COmpressible COnvection in a BOx of $L$ Dimensions, $L=2,3$ Ó)."655" It uses operator splitting (?) to separate the various (explicit) operators: the hydrodynamics, the optional tensor viscosity, and the radiation transport."," It uses operator splitting \citep{1968SJNA....5..506S} to separate the various (explicit) operators: the hydrodynamics, the optional tensor viscosity, and the radiation transport."656 The hydrodynamics module is based on a finite volume approach and relies on directional splitting to reduce the 2D or 3D problem to one dimension., The hydrodynamics module is based on a finite volume approach and relies on directional splitting to reduce the 2D or 3D problem to one dimension.657" In the 1D steps an approximate Riemann solver of Roe-type (?) is applied, modified to account for a realistic equation of state, a non-equidistant Cartesian grid, and the presence of source terms due to an external gravity field."," In the 1D steps an approximate Riemann solver of Roe-type \citep{1986AnRFM..18..337R} is applied, modified to account for a realistic equation of state, a non-equidistant Cartesian grid, and the presence of source terms due to an external gravity field."658" In addition to the stabilizing mechanism inherent in an upwind-scheme with a monotonic reconstruction method (typically a piecewise-linear van Leer interpolation), a 2D or 3D tensor viscosity can be activated."," In addition to the stabilizing mechanism inherent in an upwind-scheme with a monotonic reconstruction method (typically a piecewise-linear van Leer interpolation), a 2D or 3D tensor viscosity can be activated."659 This step eliminates certain errors of Godunov-type methods dealing with strong velocity fields aligned with the grid (?).., This step eliminates certain errors of Godunov-type methods dealing with strong velocity fields aligned with the grid \citep{1994IJNMF..18..555Q}.660" The equation of state uses pre-tabulated values as functions of density and internal energy (o,e;>P,T,s)."," The equation of state uses pre-tabulated values as functions of density and internal energy $\left(\rho,e_i\rightarrow P,\Gamma_1,T,s\right)$."661" It accounts forHI,HII, H2, Hel,Hell,HellI and a representative metal for any prescribed chemical composition."," It accounts for, $_{2}$ , and a representative metal for any prescribed chemical composition."662" The equation of state does not account for the ionization states of metals, but it uses only one neutral element to achieve the appropriate atomic weight (in the neutral case) for a given composition."," The equation of state does not account for the ionization states of metals, but it uses only one neutral element to achieve the appropriate atomic weight (in the neutral case) for a given composition."663" Two different geometries can be used with CO?BOLD, that are characterized by different gravitational potentials, boundary conditions, and modules for the radiation transport: As the outer boundaries are usually either hit at some angle by an outgoing shockwave, or let material fall back (mostly with supersonic velocities), there is not much point in tuning the formulation for an optimum transmission of small-amplitude waves."," Two different geometries can be used with $^{5}$ BOLD, that are characterized by different gravitational potentials, boundary conditions, and modules for the radiation transport: As the outer boundaries are usually either hit at some angle by an outgoing shockwave, or let material fall back (mostly with supersonic velocities), there is not much point in tuning the formulation for an optimum transmission of small-amplitude waves."664" Instead, a simple and stable prescription, that lets the shocks pass, is sufficient."," Instead, a simple and stable prescription, that lets the shocks pass, is sufficient."665 It is implemented by filling typically two layers of ghost cells where the velocity components and the internal energy are kept constant., It is implemented by filling typically two layers of ghost cells where the velocity components and the internal energy are kept constant.666" The density is assumed to decrease exponentially in the ghost layers, with a scale height set to a controllable fraction of the local hydrostatic pressure scale height."," The density is assumed to decrease exponentially in the ghost layers, with a scale height set to a controllable fraction of the local hydrostatic pressure scale height."667 The control parameter allows to account for the fact that the turbulent pressure plays a significant role for the average pressure stratification., The control parameter allows to account for the fact that the turbulent pressure plays a significant role for the average pressure stratification.668 The acceleration due to gravity is derived from Eq. 4.., The acceleration due to gravity is derived from Eq. \ref{eqgrav}.669 Within a radius rg the potential is smoothed (Eq. 4))., Within a radius $r_0$ the potential is smoothed (Eq. \ref{eqgrav}) ).670 In this sphere a source term to the internal energy provides the stellar luminosity., In this sphere a source term to the internal energy provides the stellar luminosity.671 Motions in the core are dampedby a drag force tosuppress dipolar oscillations., Motions in the core are dampedby a drag force tosuppress dipolar oscillations.672 The hydrodynamics and the radiation transport scheme ignore the core completely and integrate right through it., The hydrodynamics and the radiation transport scheme ignore the core completely and integrate right through it.673The code is parallelized with Open Multi-Processing (OpenMP) directives.,The code is parallelized with Open Multi-Processing (OpenMP) directives.674which is reduced to 3000.,which is reduced to 3000.675" The low angular frequency cut-offs for both two- and three-point statistics are meant to exclude the deeply non-linear clustering regime and thus improve the simplistic approximations for the covariances, particularly for the bispectrum."," The low angular frequency cut-offs for both two- and three-point statistics are meant to exclude the deeply non-linear clustering regime and thus improve the simplistic approximations for the covariances, particularly for the bispectrum."676" Additionally, we combine the constraints from two- and three-point statistics by simply adding Fisher matrices or multiplying posteriors, respectively, i.e. we assume that power spectra and bispectra are uncorrelated (which, again, is simplistic but common practice, e.g. 2010)."," Additionally, we combine the constraints from two- and three-point statistics by simply adding Fisher matrices or multiplying posteriors, respectively, i.e. we assume that power spectra and bispectra are uncorrelated (which, again, is simplistic but common practice, e.g. )."677" In Fig.6 we contrast the parameter constraints in the Qm—og plane from the standard Fisher matrix and the Box-Cox-Fisher analysis for two-point statistics, three-point statistics, and both data sets combined."," In $\,$ we contrast the parameter constraints in the $\Omega_{\rm m}-\sigma_8$ plane from the standard Fisher matrix and the Box-Cox-Fisher analysis for two-point statistics, three-point statistics, and both data sets combined."678" The posterior for the bispectrum constraints alone also features the characteristic Qm—og degeneracy, albeit with a tilted degeneracy line, a property that is also captured by the standard Fisher matrix (see2010)."," The posterior for the bispectrum constraints alone also features the characteristic $\Omega_{\rm m}-\sigma_8$ degeneracy, albeit with a tilted degeneracy line, a property that is also captured by the standard Fisher matrix <cit.>[see."679" Since the intersection of the contours is at a sufficiently large angle, the joint constraints in the Box-Cox-Fisher case produce fairly elliptical confidence contours which are of similar size as those resulting from the standard Fisher matrix."," Since the intersection of the contours is at a sufficiently large angle, the joint constraints in the Box-Cox-Fisher case produce fairly elliptical confidence contours which are of similar size as those resulting from the standard Fisher matrix."680 The marginalised constraints on Qm and og presented in Table allow for a more quantitative evaluation., The marginalised constraints on $\Omega_{\rm m}$ and $\sigma_8$ presented in Table allow for a more quantitative evaluation.681 Taking into account the accurate shape of the posterior generally increases the 2σ confidence interval substantially., Taking into account the accurate shape of the posterior generally increases the $2\sigma$ confidence interval substantially.682" This increase is stronger the greater the deviation of the posterior from a Gaussian shape, see e.g. the increase by 50% for Qm in the power spectrum analysis."," This increase is stronger the greater the deviation of the posterior from a Gaussian shape, see e.g. the increase by $50\,\%$ for $\Omega_{\rm m}$ in the power spectrum analysis."683" In the case of the joint two- and three-point constraints, the absolute change in errors is smaller, but still the 2c confidence interval grows by about 40% (30%) for og (Qa)."," In the case of the joint two- and three-point constraints, the absolute change in errors is smaller, but still the $2\sigma$ confidence interval grows by about $40\,\%$ $30\,\%$ ) for $\sigma_8$ $\Omega_{\rm m}$ )."684 In this work we introduced a novel method to compute precise predictions for statistical constraints on model parameters from future experiments., In this work we introduced a novel method to compute precise predictions for statistical constraints on model parameters from future experiments.685" By combining two generic statistical tools — the Fisher matrix and Box-Cox transformations, we were able to drop the assumption of Gaussianity in parameter space."," By combining two generic statistical tools – the Fisher matrix and Box-Cox transformations, we were able to drop the assumption of Gaussianity in parameter space."686" Applying Box-Cox transformations to model parameters, one arrives at approximately multivariate Gaussian shapes of the posterior."," Applying Box-Cox transformations to model parameters, one arrives at approximately multivariate Gaussian shapes of the posterior."687 In this transformed space the Fisher matrix can be computed without suffering from the usual limits of the Gaussian assumption., In this transformed space the Fisher matrix can be computed without suffering from the usual limits of the Gaussian assumption.688 An inverse Box-Cox transformation of the Fisher matrix results then yields realistic posterior distributions in the original parameter space., An inverse Box-Cox transformation of the Fisher matrix results then yields realistic posterior distributions in the original parameter space.689 We derived the formalism of the combined Fisher and Box-Cox analysis and detailed different approaches to determining the parameters of the Box-Cox transformation from an inital likelihood analysis., We derived the formalism of the combined Fisher and Box-Cox analysis and detailed different approaches to determining the parameters of the Box-Cox transformation from an inital likelihood analysis.690" Utilising a mock weak lensing survey, we verified the accuracy of the Box- formalism and demonstrated that it robustly accounts for changes in various survey parameters and"," Utilising a mock weak lensing survey, we verified the accuracy of the Box-Cox-Fisher formalism and demonstrated that it robustly accounts for changes in various survey parameters and"691he outer superadiabatic laver aud the stellar interior enough tine to thermally readjust.,the outer superadiabatic layer and the stellar interior enough time to thermally readjust.692 The superadiabatic aver progressively recoustructs aud survives the eutire evolution. making the evolution not adiabatic both ocally and eloballv.," The superadiabatic layer progressively reconstructs and survives the entire evolution, making the evolution not adiabatic both locally and globally."693 For high amass loss rates (M3L1 M.//vr)h the outer part of the star progressivelv oscs its superadiabaticitv and the interior docs not ive enough time to thermally readjust.," For high mass loss rates $\dot{M} \geq 0.1$ /yr), the outer part of the star progressively loses its superadiabaticity and the interior does not have enough time to thermally readjust."694 Even though a raction of the iuitial superadiabatic laver might survive. a larecr radiative zone emerges below it and the star seeps shrinking during the eutire sequence.," Even though a fraction of the initial superadiabatic layer might survive, a larger radiative zone emerges below it and the star keeps shrinking during the entire sequence."695 The evolution of the star is locally non-adiabatic aud hivdrodyvuiuauic. as sole enerev that is stored in gravitational form iu the wdrostatic models is actually in a kinetic form. leading o the star contracting instead of expanding.," The evolution of the star is locally non-adiabatic and hydrodynamic, as some energy that is stored in gravitational form in the hydrostatic models is actually in a kinetic form, leading to the star contracting instead of expanding."696 We also carried out additional simulations for a AAGD star with a core mass of aud a5 RRGD star., We also carried out additional simulations for a AGB star with a core mass of and a RGB star.697 These models are cousisteut with our xevious findings aud with the RRGB model in Woods&Ivanova(2011)., These models are consistent with our previous findings and with the RGB model in \cite{WoodsIvanova2011}.698. We lave also verified that the outcomes of our simulations depend ou wither nunerical parameters such as the initial timestep adopted. nor on boundary conditions.," We have also verified that the outcomes of our simulations depend on neither numerical parameters such as the initial timestep adopted, nor on boundary conditions."699 According to our stellar evolution models. eiauts barely expand. if at all.," According to our stellar evolution models, giants barely expand, if at all."700 This result nuüpacts the condition for the onset of the common envelope phase., This result impacts the condition for the onset of the common envelope phase.701 Using the Eddington huuinositv limit. one cau estimate the mass loss rate above which a dwarf would be unable to accrete material. to be about 107 ML.//vr.," Using the Eddington luminosity limit, one can estimate the mass loss rate above which a dwarf would be unable to accrete material, to be about $10^{-3}$ /yr."702 For higher mass loss rates. the hydrostatic assumption is violated aud there is no expansion of the eiant’s cuvelope.," For higher mass loss rates, the hydrostatic assumption is violated and there is no expansion of the giant's envelope."703 As a consequence. the positive feedback from the mass-losing eiut discussed in Section 1.. may be reduced.," As a consequence, the positive feedback from the mass-losing giant discussed in Section \ref{sec:intro}, may be reduced."704 Further investigations are required to quautify how this feedback affects the temporal evolution of the mass transfer rate., Further investigations are required to quantify how this feedback affects the temporal evolution of the mass transfer rate.705 Overall. criteria for unstable mass transfer based on adiabatic mass loss models should be re-investigated.," Overall, criteria for unstable mass transfer based on adiabatic mass loss models should be re-investigated."706 Moreover. if eiat stars do not expand as a result of lass loss. this process docs also not help the envelope ejection durinuga common euvelope interaction. as speculated by DeMarcoetal.(2011).," Moreover, if giant stars do not expand as a result of mass loss, this process does also not help the envelope ejection during a common envelope interaction, as speculated by \cite{AlphaPaper2011}."707. J-CP acknowledges fuudius frou NSF evant 0607111 and thanks MordecaiMark Mac Low for his support., J-CP acknowledges funding from NSF grant 0607111 and thanks Mordecai-Mark Mac Low for his support.708 FIL acknowledges finding from an NSERC Discovery eraut., FH acknowledges funding from an NSERC Discovery grant.709 BP acknowlecdecs funding οι NSF erauts PHY H-HALIGL and AST 07-07633., BP acknowledges funding from NSF grants PHY 05-51164 and AST 07-07633.710 The authors are erateful o Orsola De Marco for the initial discussions reearding he stellar response to commnon-envelope-incdiuced niass oss., The authors are grateful to Orsola De Marco for the initial discussions regarding the stellar response to common-envelope-induced mass loss.711 J-CP thanks Aaron Dotter for his help installing heFrecEOS tables. aud Charl Sakari for proofreading his mnamuscript.," J-CP thanks Aaron Dotter for his help installing the tables, and Charli Sakari for proofreading this manuscript."712"good accuracy: for €—610 (33)) gives at~0.03430 with accuracy Aa,fa,=107.","good accuracy: for $\epsilon=6\cdot 10^{-5}$ \ref{roota10}) ) gives $a^{appr}_1\simeq7130.03430$ with accuracy $\Delta a_1/a_1=10^{-3}$."714 Velocity of the wind at the critical point is approximately B30ay (o=ay τας].," Velocity of the wind at the critical point is approximately $30\,715a_s$ $v_{c}=a_1^{-1} a_s$ )."716 Above the critical point the influence of the do/dr (GEF) on the additional acceleration of (he racdiationally driven wind is small compared to the effect due to dv/dr., Above the critical point the influence of the $d\phi/dr$ (GEF) on the additional acceleration of the radiationally driven wind is small compared to the effect due to $dv/dr$.717 The most successive model ceseribing winds from hot stars presented. so [ar is that of Castor (1970). Castor.Abbott Ixlein (1975).," The most successive model describing winds from hot stars presented so far is that of Castor (1970), Castor,Abbott Klein (1975)."718 Geometry. and jonizaiion balance dilfer O-star wind from AGN outflow., Geometry and ionization balance differ O-star wind from AGN outflow.719 Line driven wind [rom O-tvpe star is spherically-svnunetric., Line driven wind from O-type star is spherically-symmetric.720 Outflows in AGN are assumed to be originated [rom the luminous accretion discs and approximately svinmetric., Outflows in AGN are assumed to be originated from the luminous accretion discs and approximately axially-symmetric.721 Powerhu X-ray and UV radiation from the disc pose a problem οἱ overionization of the outIlowing plasma., Powerful X-ray and UV radiation from the disc pose a problem of overionization of the outflowing plasma.722 In Sobolev approximation. which is in the background of the CAI theory. the raclially streaming radiation [τιν is absorbed in a line transition in a wind with the gradually increasing velocity.," In Sobolev approximation, which is in the background of the CAK theory, the radially streaming radiation flux is absorbed in a line transition in a wind with the gradually increasing velocity."723 A photon. emitted by a dise will be red-shifted due to Doppler effect.," A photon, emitted by a disc will be red-shifted due to Doppler effect."724 The resultant Sobolev optical depth τι~(de/dr)5., The resultant Sobolev optical depth $\tau_l\sim(dv/dr)^{-1}$.725" As it was shown by CAK. (he radiation force that is due to an ensemble of optically thin and optically (hick lines is proportional to (doe/dr)""."," As it was shown by CAK, the radiation force that is due to an ensemble of optically thin and optically thick lines is proportional to $(dv/dr)^\alpha$."726 The more de/dr the more effectively a line is shifted to the powerful continuum., The more $dv/dr$ the more effectively a line is shifted to the powerful continuum.727" In a number of studies. CAI theory was applied to AGN in order to explain [ast (up to 0,10) outflows Arav Li (1994). Arav. Li Degelman (1994). Murray. at al. ("," In a number of studies, CAK theory was applied to AGN in order to explain fast (up to $\sim 0.1 c$ ) outflows Arav Li (1994), Arav, Li Begelman (1994), Murray at al. ("7281995). Proga et al. (,"1995), Proga et al. ("7291993). Proga et al. (,"1998), Proga et al. ("7302000).,2000).731 The CAI theory was enhanced by adopting axial svmnieltry., The CAK theory was enhanced by adopting axial symmetry.732 Ionization balance was studied in details simultaneously with 2D hydrodvimanical ealeulations., Ionization balance was studied in details simultaneously with 2D hydrodymanical calculations.733 In this paper we developed a theorv of winds (hat takes into account effects of the strong gravitational fields., In this paper we developed a theory of winds that takes into account effects of the strong gravitational fields.734" We point out that if a wind is accelerated near super-massive DII a gravitational change of photon's Irequencey"" must be taken into account.", We point out that if a wind is accelerated near super-massive BH a gravitational change of photon's frequency must be taken into account.735 Ii a strong gravitational field a photon. emitted bv a disc. will be red-shifted. due to both the Doppler effect ~c/c and gravitational redshifting Ανν=.No/c.," In a strong gravitational field a photon, emitted by a disc, will be red-shifted due to both the Doppler effect $\sim v/c$ and gravitational redshifting $\Delta\nu/\nu=\Delta\phi/c^2$."736 We argue that taking into account gravitational redshifling can substantially change the wind dvnamics and structure., We argue that taking into account gravitational redshifting can substantially change the wind dynamics and structure.737 Although it should be mentioned that the developed theory (in the adopted tov model) cannot be directly applied to explain outflows from close to. DII., Although it should be mentioned that the developed theory (in the adopted toy model) cannot be directly applied to explain outflows from close to BH.738 An exact self-consistent solution of this problem is possible onlv in general relativity., An exact self-consistent solution of this problem is possible only in general relativity.739 We avoid this sophisticated task by considering all equations in flat space and time., We avoid this sophisticated task by considering all equations in flat space and time.740"There is a theoretical dilemma concerning Jupiter formation, as described in Introduction.","There is a theoretical dilemma concerning Jupiter formation, as described in Introduction."741" Modelings of Jupiter's interior suggest that Jupiter has a small core of « 10Ms (e.g.,Saumon&Guillot2004),, while many core-accretion models of Jupiter formation require alarge core of > 10Ma to finish its formation by the time of disk dissipation (Pollacketal.1996;Alibert2004,2005;Fortier2007,2009)."," Modelings of Jupiter's interior suggest that Jupiter has a small core of $<$ $10 M_\oplus$ \citep[e.g.,][]{sg04}, while many core-accretion models of Jupiter formation require alarge core of $>$ $10 M_\oplus$ to finish its formation by the time of disk dissipation \citep[][]{jbp96,alibert04,alibert05,fortier07,fortier09}."742. The disk instability scenario has been revisited as an alternative scenario of their formation 2002).," The disk instability scenario has been revisited as an alternative scenario of their formation \citep[e.g.][]{boss00,mayer02}."743" In this study, we have demonstrated that reduced opacities in the protoplanet's envelope have the potential to resolve this dilemma."," In this study, we have demonstrated that reduced opacities in the protoplanet's envelope have the potential to resolve this dilemma."744" From Fig. 2,,"," From Fig. \ref{fig2},"745 one finds that Moore=0.75Ma for το= 1 Myr in the metal-free case; Moore=1.7Ms even in the alkali case., one finds that $M_{\rm core} = 0.75 M_\oplus$ for $\tau_g =$ 1 Myr in the metal-free case; $M_{\rm core} = 1.7 M_\oplus$ even in the alkali case.746" Given observed lifetimes of protoplanetary disks of several Myr, the fact above indicates the reduction of opacity allows Jupiter to have a small core that is consistent with interior modelings in principle."," Given observed lifetimes of protoplanetary disks of several Myr, the fact above indicates the reduction of opacity allows Jupiter to have a small core that is consistent with interior modelings in principle."747" The feasibility of such minimum M, depends on opacities in the protoplanet’s envelope, while this does not change our conclusion that the minimum Mj, obtained here provides the lowest limit to core masses of gas giants to which the core accretion model can apply."," The feasibility of such minimum $M_\mathrm{crit}$ depends on opacities in the protoplanet's envelope, while this does not change our conclusion that the minimum $M_\mathrm{crit}$ obtained here provides the lowest limit to core masses of gas giants to which the core accretion model can apply."748" We need more extensive investigation of gas giant formation into which sedimentation and coagulation of grains in the accreting envelope are incorporated, although reduction of opacity was already pronounced (Podolak2003;Movshovitz&Podolak2008)."," We need more extensive investigation of gas giant formation into which sedimentation and coagulation of grains in the accreting envelope are incorporated, although reduction of opacity was already pronounced \citep{mp03,mp08}."749. Our results also shed light upon growth of solid cores., Our results also shed light upon growth of solid cores.750 Many core accretion models assumed the presence of a single protoplanet (e.g.Pollacketal.1996)., Many core accretion models assumed the presence of a single protoplanet \citep[e.g.][]{jbp96}.751". However, a gas giant is in practice thought of as being formed in a system of multiple protoplanets embedded in a protoplanetary disk."," However, a gas giant is in practice thought of as being formed in a system of multiple protoplanets embedded in a protoplanetary disk."752" Compared to cases of a single protoplanet, the final mass of a core should be small in the case of a multiple-protoplanet system."," Compared to cases of a single protoplanet, the final mass of a core should be small in the case of a multiple-protoplanet system."753" According to 2000),, the isolation mass is a few Ms around 5AU."," According to \citet{ki98,ki00}, , the isolation mass is a few $M_\oplus$ around $5$ AU."754 Even such a small core is enough, Even such a small core is enough755"and ages of individual sources. and hence the jet ""on"" time as a function of stellar mass.","and ages of individual sources, and hence the jet “on” time as a function of stellar mass."756 The bivariate luminosity function was then used to constrain the time a twpical radio source spends in an inactive state., The bivariate luminosity function was then used to constrain the time a typical radio source spends in an inactive state.757 Hadio and emission line AGN activity are found to be independent phenomena., Radio and emission line AGN activity are found to be independent phenomena.758 We also find that both the radio source lifetime and. duration of the quiescent phase have a strong mass dependence. with massive hosts harbouring longcr-lived sources that are triggered more frequently.," We also find that both the radio source lifetime and duration of the quiescent phase have a strong mass dependence, with massive hosts harbouring longer-lived sources that are triggered more frequently."759 Cas cooling rate shows a similar mass dependence. suggesting that fuel depletion is the reason the jets switch oll.," Gas cooling rate shows a similar mass dependence, suggesting that fuel depletion is the reason the jets switch off."760 We thank the Commonwealth. Cambridge Trust. (S8) and he Isaac Newton ‘Trust (SS and. SA) for support. and the anonymous referee for comments that have helped improve he paper.," We thank the Commonwealth Cambridge Trust (SS) and the Isaac Newton Trust (SS and SA) for support, and the anonymous referee for comments that have helped improve the paper."761 This work makes use of the SDSS Archive. πο for which has been provided by the Alfred. P. Sloan Foundation. the Participating Institutions. the National Science. Foundation. the U.S. Department of Energy. he National Acronautics ancl Space Administration. the Japanese Monbukagakusho. the Max. Planck Society. and he Higher Education Funding Council lor England.," This work makes use of the SDSS Archive, funding for which has been provided by the Alfred P. Sloan Foundation, the Participating Institutions, the National Science Foundation, the U.S. Department of Energy, the National Aeronautics and Space Administration, the Japanese Monbukagakusho, the Max Planck Society, and the Higher Education Funding Council for England."762 This work also makes use of the NVSS and FIRST surveys carried out using the National Radio Astronomy Observatory Very Large Array., This work also makes use of the NVSS and FIRST surveys carried out using the National Radio Astronomy Observatory Very Large Array.763 The National Racio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities. Inc.," The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc."764Therefore. it is quite important to gain physical insight into how the radiation of the first stars impacted their surroundings.,"Therefore, it is quite important to gain physical insight into how the radiation of the first stars impacted their surroundings."765 Radiative feedback can be either positive or negative. in that it can enhance or suppress subsequent star-formation.," Radiative feedback can be either positive or negative, in that it can enhance or suppress subsequent star–formation."766 Positive feedback can result when the enhanced free-electron fraction from ionizing photons (e.g. 2)) or hydrodynamical shocks (2) catalyzes the formation of molecular hydrogen (H2).," Positive feedback can result when the enhanced free-electron fraction from ionizing photons (e.g. \citealt{OH02}) ) or hydrodynamical shocks \citep{SK87}767 catalyzes the formation of molecular hydrogen $_2$ )."768" If the ionizing background ""turns-off (resulting in so-called ""relie"" HIT regions). H» cooling can provide the dominant cooling channel at high-densities and low temperatures."," If the ionizing background “turns-off"" (resulting in so-called “relic"" HII regions), $_2$ cooling can provide the dominant cooling channel at high-densities and low temperatures."769 Conversely. negative feedback can result from heating by ionizing radiation which can photo-evaporate gas in low-mass halos (222222).," Conversely, negative feedback can result from heating by ionizing radiation which can photo-evaporate gas in low-mass halos \citep{Efstathiou92, BL99, Gnedin00filter, SIR04, Dijkstra04, MD08}."770 Also. an active background of Lyman-Werner (LW) radiation (with photon energies in the |1.18—13.6eV. range) can dissociate Πο. thus decreasing the gas’s cooling capabilities (e.g... 222222).," Also, an active background of Lyman-Werner (LW) radiation (with photon energies in the 11.18–13.6eV range) can dissociate $_2$, thus decreasing the gas's cooling capabilities (e.g., \citealt{HRL97, HAR00, CFA00, MBA01, WA07, Oshea08}) )."771 Indeed. simulations find radiative feedback to be nuanced at very high redshifts.," Indeed, simulations find radiative feedback to be nuanced at very high redshifts."772 Positive feedback dominates in flash ionized gas (2).. or gas exposed to a weak transient ultraviolet background (UVB) (2.. hereafter MBHO06) such as might be present close to the edges of HII regions (22)..," Positive feedback dominates in flash ionized gas \citep{OShea05}, or gas exposed to a weak transient ultraviolet background (UVB) \citealt{MBH06}, hereafter MBH06) such as might be present close to the edges of HII regions \citep{RGS02b, KM05}."773 On the other hand. negative feedback in relic HII regions can occur in regions closer to the Pop III star (MBHO06: ???)).," On the other hand, negative feedback in relic HII regions can occur in regions closer to the Pop III star (MBH06; \citealt{SU06, AS07, Yoshida07}) )."774 Furthermore. radiative transfer effects can also impact the strength and sign of feedback (e.g. 22222).," Furthermore, radiative transfer effects can also impact the strength and sign of feedback (e.g. \citealt{ISR05, SU06, AWB07, Whalen08, WA08}) )."775 Because of variations in halo mass. collapse redshift. central baryon density. stellar spectrum. and distance to the ionizing source. it is difficult to accurately deal with the large parameter space governing photoevaporation.," Because of variations in halo mass, collapse redshift, central baryon density, stellar spectrum, and distance to the ionizing source, it is difficult to accurately deal with the large parameter space governing photoevaporation."776 In MBH06. we attempted to tackle this issue with a statistica approach.," In MBH06, we attempted to tackle this issue with a statistical approach."777" By giving up on simulating ""realistic? (as much as this is possible given present uncertainties) relic HII regions left behind by deceased Pop III stars. we were able to explore larger swaths of parameter space and have a fairly large sample of second-generation objects."," By giving up on simulating “realistic"" (as much as this is possible given present uncertainties) relic HII regions left behind by deceased Pop III stars, we were able to explore larger swaths of parameter space and have a fairly large sample of second-generation objects."778 We did this by applying various uniform ultraviolet (UV) and LW backgrounds on our entire simulation box. and statistically comparing the resulting evolution agains the fiducial run without radiation.," We did this by applying various uniform ultraviolet (UV) and LW backgrounds on our entire simulation box, and statistically comparing the resulting evolution against the fiducial run without radiation."779 Although this approach clearly does not realistically treat the photoionization around a particular source. it might provide a good statistical description of a ensemble of reionization studies.," Although this approach clearly does not realistically treat the photoionization around a particular source, it might provide a good statistical description of a ensemble of reionization studies."780 Clearly. at the onset of early reionization not all halos lie within HII regions. as they do in our models.," Clearly, at the onset of early reionization not all halos lie within HII regions, as they do in our models."781 Our primary purpose is not to simulate early reionization or the photoevaporation of well-formed Pop III halos in detail (which would require radiative transfer) but to evaluate the impact of relic HII regions on the formation of halos at later times., Our primary purpose is not to simulate early reionization or the photoevaporation of well-formed Pop III halos in detail (which would require radiative transfer) but to evaluate the impact of relic HII regions on the formation of halos at later times.782 Our uniformly illuminated boxes provide an ensemble of evaporated halos at high redshift that statistically sample the effects of early photoionization on the assembly of halos and formation of cold dense gas at lower redshifts., Our uniformly illuminated boxes provide an ensemble of evaporated halos at high redshift that statistically sample the effects of early photoionization on the assembly of halos and formation of cold dense gas at lower redshifts.783 However. since our analysis in MBH06 focused on a highly biased region which went non-linear at a high redshift. we could not follow the evolution to redshifts much lower than 2.20.," However, since our analysis in MBH06 focused on a highly biased region which went non-linear at a high redshift, we could not follow the evolution to redshifts much lower than $z\sim20$."784 This meant that we could only extrapolate the observed trends suggesting that the feedback was transient., This meant that we could only extrapolate the observed trends suggesting that the feedback was transient.785 ? claim that. even in the presence of a weak LWB. excess entropy will eventually suppress star formation in protogalaxies forming inside relic HIT regions.," \citet{OH03} claim that, even in the presence of a weak LWB, excess entropy will eventually suppress star formation in protogalaxies forming inside relic HII regions."786 In MBHO06. we were unable to conclusively verify or refute this claim.," In MBH06, we were unable to conclusively verify or refute this claim."787 In this work. we extend the study by MBHO06 by analyzing radiative feedback in a less-biased region. whose typical proto-galaxies form at lower redshifts. 2~ 13—20.," In this work, we extend the study by MBH06 by analyzing radiative feedback in a less-biased region, whose typical proto-galaxies form at lower redshifts, $z\sim$ 13–20."788 Given that the 5-yr WMAP polarization data place a constraint on the reionization redshift (assuming instantaneous reionization) of +=11.0c1.4 (2).. such cosmological regions are likely to host the majority of feedback effects from Pop III stars. before a persistent UVB and/or a strong LWB become entrenched (e.g. 2)).," Given that the 5-yr polarization data place a constraint on the reionization redshift (assuming instantaneous reionization) of $z=11.0 \pm 1.4$ \citep{Dunkley08}, such cosmological regions are likely to host the majority of feedback effects from Pop III stars, before a persistent UVB and/or a strong LWB become entrenched (e.g. \citealt{HRL96}) )."789 We apply the same statistical approach as in MBHO6. and study various combinations of transient UVBs and persistent LWBs.," We apply the same statistical approach as in MBH06, and study various combinations of transient UVBs and persistent LWBs."790 As such. we attempt to provide a framework for analytically incorporating such radiative feedback into semi-analytic. and large-scale numerical and semi-numerical studies.," As such, we attempt to provide a framework for analytically incorporating such radiative feedback into semi-analytic, and large-scale numerical and semi-numerical studies."791 In $2. we enumerate and describe the simulations which are used in this work., In \ref{sec:sims} we enumerate and describe the simulations which are used in this work.792 In $3.. we study the initial radiative feedback caused by a transient UVB. adding a persistent LWB in $3.1..," In \ref{sec:trans}, we study the initial radiative feedback caused by a transient UVB, adding a persistent LWB in \ref{sec:LW}."793 Then in S4+.. we describe a related mechanism that results in eventual positive feedback. regardless of the strength of the UVB.," Then in \ref{sec:pos}, we describe a related mechanism that results in eventual positive feedback, regardless of the strength of the UVB."794 In discuss the impact of neglecting self-shielding.," In \ref{sec:ss}, we discuss the impact of neglecting self-shielding."795 Finally. we offer our conclusions in S6..," Finally, we offer our conclusions in \ref{sec:conc}."796 Throughout this paper. we adopt the background cosmological parameters (Ox. Ox; Ou. n. σκι Ho) = (0.7. 0.3. 0.047. 1. 0.92. 70 km + +). consistent with the measurements of the power spectrum of CMB temperature anisotropies by the first vear of data from the satellite (2)...," Throughout this paper, we adopt the background cosmological parameters $\Omega_\Lambda$, $\Omega_{\rm M}$, $\Omega_b$, n, $\sigma_8$, $H_0$ ) = (0.7, 0.3, 0.047, 1, 0.92, 70 km $^{-1}$ $^{-1}$ ), consistent with the measurements of the power spectrum of CMB temperature anisotropies by the first year of data from the satellite \citep{Spergel03}."797 Although the S-yr data prefers a slightly lower value of σε=0.52 (22). which somewhat delays structure formation. we keep the cosmology the same as in MBHO6. to facilitate direct comparison.," Although the 5-yr data prefers a slightly lower value of $\sigma_8=0.82$ \citep{Dunkley08, Komatsu08} which somewhat delays structure formation, we keep the cosmology the same as in MBH06, to facilitate direct comparison."798 Unless stated otherwise. we quote all quantities in comoving units.," Unless stated otherwise, we quote all quantities in comoving units."799 We use the Eulerian adaptive mesh refinement (AMR) code Enzo. which is described in greater detail elsewhere (22)..," We use the Eulerian adaptive mesh refinement (AMR) code Enzo, which is described in greater detail elsewhere \citep{Bryan99, NB99}."800" Our simulation volume is | (5.!Mpe)*. initialized at zi,=99 with density perturbations drawn from the ?. power spectrum."," Our simulation volume is 1 $(\hMpc)^3$, initialized at $\zinit=99$ with density perturbations drawn from the \citet{EH99} power spectrum."801" Our root grid is 1283,", Our root grid is $128^3$.802 We have two additional static levels of retinement inside a central 0.25 .!Alpe cube., We have two additional static levels of refinement inside a central 0.25 $\hMpc$ cube.803 In addition. grid cells inside the central region are allowed to dynamically retine so that the Jeans length is resolved by at least + grid zones and no grid cell contains more than 4 times the initial gas mass element.," In addition, grid cells inside the central region are allowed to dynamically refine so that the Jeans length is resolved by at least 4 grid zones and no grid cell contains more than 4 times the initial gas mass element."804 Each additional grid level refines the mesh length of the parent grid cell by a factor of 2., Each additional grid level refines the mesh length of the parent grid cell by a factor of 2.805 We allow for a maximum of IO levels of refinement inside the refined central region. granting us a spatial resolution of 7.63 f+pe.," We allow for a maximum of 10 levels of refinement inside the refined central region, granting us a spatial resolution of 7.63 $h^{-1}~{\rm pc}$."806 This comoving resolution translates to 0.36 ήproperpe at >=20., This comoving resolution translates to 0.36 $h^{-1}~{\rm proper~pc}$ at $z=20$.807 As stated above. our simulation runs were set up to facilitate comparison with MBHO06.," As stated above, our simulation runs were set up to facilitate comparison with MBH06."808 Users interested in details of the simulations are encouraged to consult MBH06 and ?.., Users interested in details of the simulations are encouraged to consult MBH06 and \citet{MBA01}.809 There is one notable difference between our runs here and those in MBH06: our central refined region for these runs is centered on the highest density region of the box., There is one notable difference between our runs here and those in MBH06: our central refined region for these runs is centered on the highest density region of the box.810 Instead is it chosen to be more typical of the regions expected to host the bulk of halos which form prior to any significant cosmological reionization., Instead is it chosen to be more typical of the regions expected to host the bulk of halos which form prior to any significant cosmological reionization.811 Specitically. the density inside our central (0.25/1Alpe)? corresponds to a 0.75 σ mass fluctuation of an equivalent spherical volume tin ΜΡΗΟΟ. we studied a 2.4 0 region).," Specifically, the density inside our central $(0.25 \hMpc)^3$ corresponds to a 0.75 $\sigma$ mass fluctuation of an equivalent spherical volume (in MBH06, we studied a 2.4 $\sigma$ region)."812 The less overdense region allows us to extend the analysis of MBH06 to lower redshifts. and test the robustness of the conclusions of MBH06 on a less-biased and more typical region.," The less overdense region allows us to extend the analysis of MBH06 to lower redshifts, and test the robustness of the conclusions of MBH06 on a less-biased and more typical region."813 As shown in Table |. we have performed four different runs without a LW background. distinguished by the duration or amplitude of the assumed UVB. and eight additional runs that include an additional constant LW background.," As shown in Table \ref{tbl:runs}, we have performed four different runs without a LW background, distinguished by the duration or amplitude of the assumed UVB, and eight additional runs that include an additional constant LW background."814 Again. these runs are analogous runs to those in MBHO6.," Again, these runs are analogous runs to those in MBH06."815 For the UV radiation we assume an isotropic background flux with a 7=2 10!K blackbody spectral shape. normalized at the hydrogen ionization frequency. rg = 13.6 eV. This spectrum is softer than the," For the UV radiation we assume an isotropic background flux with a $T=2\times10^4$ K blackbody spectral shape, normalized at the hydrogen ionization frequency, $h \nu_H$ = 13.6 eV. This spectrum is softer than the"816the corresponding apparent distribution using inversion techniques based on the assumption that their orientations are random.,the corresponding apparent distribution using inversion techniques based on the assumption that their orientations are random.817 In this paper we use the APAL catalogue (Dalton et al., In this paper we use the APM catalogue (Dalton et al.818 1997) to measure the projected. shape distribution. Corrected for various systematic ellects. and hence attempt to estimate the intrinsic shape of clusters.," 1997) to measure the projected shape distribution, corrected for various systematic effects, and hence attempt to estimate the intrinsic shape of clusters."819 The XPM clusters are typically as rich as Abell 2=0 clusters. but due to the careful identification procedure do not sulfer from significant projection ellects.," The APM clusters are typically as rich as Abell $R=0$ clusters, but due to the careful identification procedure do not suffer from significant projection effects."820 The plan of the paper is the following: In. Section 2 we describe the APM galaxy. ancl cluster survey. in Section Lx Pa oesent our projected cluster. shape determination method anc by using Monte. Carlo. simulations we establish its statistical robustness.," The plan of the paper is the following: In Section 2 we describe the APM galaxy and cluster survey, in Section 3 we present our projected cluster shape determination method and by using Monte Carlo simulations we establish its statistical robustness."821 We discuss how the foregroundbackground. Contamination (projection elfects) alfects the projected cluster shapes and present a statistical ellipticity correction procedure., We discuss how the foreground/background contamination (projection effects) affects the projected cluster shapes and present a statistical ellipticity correction procedure.822 The correction assumes that clusters are in dynamical equilibrium. and. therefore. we exclude clusters with strong. substructure., The correction assumes that clusters are in dynamical equilibrium and therefore we exclude clusters with strong substructure.823 In. Section + we invert the systematic bias-corrected projected ellipticity distribution to recover the intrinsic one., In Section 4 we invert the systematic bias-corrected projected ellipticity distribution to recover the intrinsic one.824 Our conclusions are presented in Section 5., Our conclusions are presented in Section 5.825 The APAL survey covers an area of 4300. square. degrees in the southern sky (6.< 40°) ancl contains about 2.5 million galaxies brighter than a magnitude limit of b;=20.5., The APM survey covers an area of 4300 square degrees in the southern sky $b\leq -40^\circ$ ) and contains about 2.5 million galaxies brighter than a magnitude limit of $b_{J}=20.5$.826 Details of the APAL cata can be found. in Aladclox et al. (, Details of the APM data can be found in Maddox et al. (8271990a and 19000). ancl Maddox. Efstathiou Sutherland (1996).,"1990a and 1990b), and Maddox, Efstathiou Sutherland (1996)."828 ere we present only a brief summary of the catalogue., Here we present only a brief summary of the catalogue.829 The survey was compiled from. 185 survey plates from the Ulx Schmidt. telescope. scanned. by. the Automatic Plate Measuring CXAPM) machine in Cambridge., The survey was compiled from 185 survey plates from the UK Schmidt telescope scanned by the Automatic Plate Measuring (APM) machine in Cambridge.830" The scanned region of cach late covers 5.875.8” of the sky. and since neighbouring plate centers are separated by 5° this leads to 0.8"" overlaps along plate boundaries."," The scanned region of each plate covers $5.8^{\circ} \times8315.8^{\circ}$ of the sky, and since neighbouring plate centers are separated by $5^{\circ}$ this leads to $0.8^{\circ}$ overlaps along plate boundaries."832 The data for cach plate is stored separately to. preserve. the multiple measurements in the overlap regions., The data for each plate is stored separately to preserve the multiple measurements in the overlap regions.833 Extensive internal checks and external calibration have shown that 10 plate-to-plate zero-point error has an rms of 0.06 =uagnitudes. and that large-scale photometric gradients are even smaller.," Extensive internal checks and external calibration have shown that the plate-to-plate zero-point error has an rms of 0.06 magnitudes, and that large-scale photometric gradients are even smaller."834 Phe LUCAS and CODI all-skv maps show that 1e galactic obscuration in this region of the sky is typically ).06 magnitucles introducing comparable uncertainty in the hotometry., The IRAS and COBE all-sky maps show that the galactic obscuration in this region of the sky is typically 0.06 magnitudes introducing comparable uncertainty in the photometry.835 The image profiles and. shapes were used. to Jassifv them into galaxies. stars and blended stars.," The image profiles and shapes were used to classify them into galaxies, stars and blended stars."836 Visual 'hecks and deeper CCD images show that the classification eads to galaxy samples which are complete with contamination of from non-galaxies., Visual checks and deeper CCD images show that the classification leads to galaxy samples which are complete with contamination of from non-galaxies.837 Dalton ct al (1997) applied an object cluster finding algorithm to the APAL galaxy data. and so produced a list of galaxy clusters. most of which have subsequently. been spectroscopically confirmed. as clusters.," Dalton et al (1997) applied an object cluster finding algorithm to the APM galaxy data, and so produced a list of galaxy clusters, most of which have subsequently been spectroscopically confirmed as clusters."838 Lhe cluster finding algorithm consists of two main steps: The first step uses a percolation algorithm to link all pairs of galaxies with separations <0.7 the mean inter-galaxy separation., The cluster finding algorithm consists of two main steps: The first step uses a percolation algorithm to link all pairs of galaxies with separations $< 0.7$ the mean inter-galaxy separation.839 MI mutually linked. pairs are joined. together to form. groups. and the groups with more than 20 galaxies are identified as candidate clusters.," All mutually linked pairs are joined together to form groups, and the groups with more than 20 galaxies are identified as candidate clusters."840 ln the second. step. an iterative routine is applied to each candidate to estimate the richness ancl characteristic apparent magnitude of galaxies within a search. radius of 0.75f+ Alpe.," In the second step, an iterative routine is applied to each candidate to estimate the richness and characteristic apparent magnitude of galaxies within a search radius of $0.75 \; h^{-1}$ Mpc."841 This. produced a list of 957 clusters with 2.25OL and APAL richness of more than 40 galaxies. corresponding roughly to Abell richness class 0.," This produced a list of 957 clusters with $z_{est} \mincir 0.1$ and APM richness of more than 40 galaxies, corresponding roughly to Abell richness class 0."842 The angular ciameter of the search radius is set to be consistent with the distance estimated from the apparent maenituce of galaxies in the search radius., The angular diameter of the search radius is set to be consistent with the distance estimated from the apparent magnitude of galaxies in the search radius.843 For our present analysis we cross-correlated the cluster positions with the PM galaxy survey. and for cach cluster selected all galaxies falling within a distance of 1.2h.! Alpe from the cluster center.," For our present analysis we cross-correlated the cluster positions with the APM galaxy survey, and for each cluster selected all galaxies falling within a distance of $1.2 \; h^{-1}$ Mpc from the cluster center."844 Since this is a larger radius than used in the cluster identification. some clusters near to the survey. boundaries do not have complete data over the full circle.," Since this is a larger radius than used in the cluster identification, some clusters near to the survey boundaries do not have complete data over the full circle."845 We found that 54 of the APM clusters are allected. and have simply rejected them from the sample. leaving 903 clusters which we use in our analysis.," We found that 54 of the APM clusters are affected, and have simply rejected them from the sample, leaving 903 clusters which we use in our analysis."846 In order to estimate the APAL cluster shapes we use the moments of inertia method. (ef, In order to estimate the APM cluster shapes we use the moments of inertia method (cf.847" Carter Aletealle 1980: Plionis. Barrow Frenk 1991) The galaxy equatorial positions are transformed. into an equal area coordinate system. centered on the cluster center. using: a=(Ra,Ray)cosas) and yo=0,0,4. where subscripts g ancl ef refer to galaxies and the cluster. respectively."," Carter Metcalfe 1980; Plionis, Barrow Frenk 1991) The galaxy equatorial positions are transformed into an equal area coordinate system, centered on the cluster center, using: $x =(Ra_{g}-Ra_{cl}) \times \cos(\delta_{cl})$ and $y =\delta_{g}-\delta_{cl}$, where subscripts $g$ and $cl$ refer to galaxies and the cluster, respectively."848 We then evaluate the moments: {τι=»areoaT) Ls=»usi we). din=doy»(ug. With ow; the statistical weight ofcach point.," We then evaluate the moments: $I_{11}=\sum\ w_{i}(r_{i}^{2}-x_{i}^{2})$, $I_{22}=\sum\ w_{i}(r_{i}^{2}-y_{i}^{2})$ , $I_{12}=I_{21}=-\sum\ w_{i}x_{i}y_{i}$, with $w_{i}$ the statistical weight of each point."849 Note that because the inertia tensor is symmetric we have fy2=fo)., Note that because the inertia tensor is symmetric we have $I_{12}=I_{21}$.850 Diagonalizing the Inertia Censor we obtain the eigenvalues Ay. À». from which we define he ellipticitv of the configuration under study by: ο=1Ani AL. with λιAz.," Diagonalizing the inertia tensor we obtain the eigenvalues $\lambda_{1}$, $\lambda_{2}$, from which we define the ellipticity of the configuration under study by: $\epsilon=1-\lambda_{2}/\lambda_{1}$ , with $\lambda_{1}>\lambda_{2}$."851 The corresponding eigenvectors xovide the direction of the principal axis., The corresponding eigenvectors provide the direction of the principal axis.852 This basic shape estimation method is applied using two alternative methods: (1) Discrete case (ee= 1): In which we use the individual galaxies to determine the cluster shape., This basic shape estimation method is applied using two alternative methods: (1) Discrete case $w=1$ ): In which we use the individual galaxies to determine the cluster shape.853 € first all galaxies within a small radius from the cluster center. (~M1ht Alpe) are used to celine the initial value of the cluster shape parameters. then the next nearest. galaxy is added: consecutively to the initial group and the shape is reealeulatecl until we include all galaxies within a limiting radius of our choice (usually ~0.75f+ Alpe). (," At first all galaxies within a small radius from the cluster center $\sim 0.1 \; h^{-1}$ Mpc) are used to define the initial value of the cluster shape parameters, then the next nearest galaxy is added consecutively to the initial group and the shape is recalculated until we include all galaxies within a limiting radius of our choice (usually $\sim 0.75 \; h^{-1}$ Mpc). ("8542) Smooth case (te= 0): In which we use the smoothed,2) Smooth case $w=\delta$ ): In which we use the smoothed855There are several phenomenological models for QCD that incorporate confinement.,There are several phenomenological models for QCD that incorporate confinement.856 The simplest of these is the hard wall model., The simplest of these is the hard wall model.857 A better one is the soft wall moclel as it represents (he linear radial Regge trajectories lor the [Families of p and αι mesons quite well., A better one is the soft wall model as it represents the linear radial Regge trajectories for the families of $\rho$ and $a_1$ mesons quite well.858" IIere we compute the ratio of shear viscosity to entropy. density aud the shear relaxation time for these models since thev may be useful in applications to high energy nuclear collisions at ΠΕΙ and LHC,", Here we compute the ratio of shear viscosity to entropy density and the shear relaxation time for these models since they may be useful in applications to high energy nuclear collisions at RHIC and LHC.859 The hard wall model [17] uses the metric of Eq. (4)), The hard wall model \cite{hardwall} uses the metric of Eq. \ref{SUSY}) )860 but with r-space cutoff at some value ry., but with $r$ -space cutoff at some value $r_{\rm min}$.861 This leads to radial excitations of the vector ancl axial-vector meson spectra with the mass being linear in (he radial quantum number 7., This leads to radial excitations of the vector and axial-vector meson spectra with the mass being linear in the radial quantum number $n$.862 Introduction of the new scale μμ implies that the transport coefficients do not scale dimensionallv with temperature 7 alone., Introduction of the new scale $r_{\rm min}$ implies that the transport coefficients do not scale dimensionally with temperature $T$ alone.863" There is a sienilicant difference depending on whether ry,<rp or ro.", There is a significant difference depending on whether $r_{\rm min} < r_0$ or $r_{\rm min} > r_0$ .864 In the former case the introduction of a hard eutoff does not influence the caleulations of the previous sections since it lies inside the black hole., In the former case the introduction of a hard cutoff does not influence the calculations of the previous sections since it lies inside the black hole.865 In the latter case it would seem to be indeterminate because the horizon lies bevond the wall and the boundary condition near the horizon cannot be implemented., In the latter case it would seem to be indeterminate because the horizon lies beyond the wall and the boundary condition near the horizon cannot be implemented.866"On the other hand. we can stretch the boundary out. from ro—rg tor=rg if the distance rg,—ry is so small that we can use the approximalions gou(r)8—5o0(r—rg) and g(r)225,/(r—ry).","On the other hand, we can stretch the boundary out from $r=r_0$ to $r=r_{\rm min}$ if the distance $r_{\rm min} - r_0$ is so small that we can use the approximations $g_{00}(r) \approx -\gamma_0 (r-r_0)$ and $g_{rr}(r) \approx \gamma_r/(r-r_0)$."867 Then the boundary condition on the stretched horizon which allows only for incoming. andnot outgoing. waves can be applied.," Then the boundary condition on the stretched horizon which allows only for incoming, andnot outgoing, waves can be applied."868 The answers are (he same as Eqs. (4)), The answers are the same as Eqs. \ref{diffusion}) )869 and (5)) with ry replaced by μι. One finds that and -—=——::lorT«T., and \ref{relaxation}) ) with $r_0$ replaced by $r_{\rm min}$ One finds that = T T_c and = T T_c870The traded spectra of the strongest emission lines are displaved. in Fig.,The trailed spectra of the strongest emission lines are displayed in Fig.871 5., 5.872 Lia shows a single-peak profile. in contrast to its double-peak profile only a month before our observations (Still et al.," $\alpha$ shows a single-peak profile, in contrast to its double-peak profile only a month before our observations (Still et al."873 997)., 1997).874 Ls flux has also increased over that. period of time. thus the couble-peak profile may still be embedded into the much stronger single-peak emission component.," Its flux has also increased over that period of time, thus the double-peak profile may still be embedded into the much stronger single-peak emission component."875 The bad CCD column. that alfects the line profile. can serve as à reference to demonstrate that the binary motion in Ho is not clearly visible. though evident in the Fourier power spectra.," The bad CCD column, that affects the line profile, can serve as a reference to demonstrate that the binary motion in $\alpha$ is not clearly visible, though evident in the Fourier power spectra."876 The LL3 profiles show more clearly a single-peak core moving with a low-amplituce velocity., The $\beta$ profiles show more clearly a single-peak core moving with a low-amplitude velocity.877 Pulses produced by the rotating beams of the white dwarf are visible on the line profiles., Pulses produced by the rotating beams of the white dwarf are visible on the line profiles.878 Phe orbital variation of the 4686. profile is complex anc very dillerent. from that of the Balmer lines., The orbital variation of the 4686 profile is complex and very different from that of the Balmer lines.879 The average profile is a double-peak profile (+400 kni s. 1)., The average profile is a double-peak profile $\pm$ 400 km $^{-1}$ ).880 Phe trailed spectra then reveal a prominent sinusoidal component which crosses from red to blue at (relative) phase 0.6 (as we will see. absolute phase 0.5) and shows maximum strength between (relative) phases 0.4-0.7.," The trailed spectra then reveal a prominent sinusoidal component which crosses from red to blue at (relative) phase 0.6 (as we will see, absolute phase 0.5) and shows maximum strength between (relative) phases 0.4-0.7."881 In. addition. another emission component is either moving in anti-phase to the previous component or is part of an asymmetric double-peak profile.," In addition, another emission component is either moving in anti-phase to the previous component or is part of an asymmetric double-peak profile."882 The IL narrow component shows a broadening of EWIIM-230628 km s which is consistent with emission from the Roche lobe of the secondary star. (Doppler-broadened. =200. Km | by the binary motion)., The narrow component shows a broadening of $\pm$ 28 km $^{-1}$ which is consistent with emission from the Roche lobe of the secondary star (Doppler-broadened $\approx200~$ km $^{-1}$ by the binary motion).883 We identify this emission. with the inner side of the red star., We identify this emission with the inner side of the red star.884 The above suggests that the inferior conjunction of the secondary star (absolute binary phase) leads by 0.1 eveles the emission-line source., The above suggests that the inferior conjunction of the secondary star (absolute binary phase) leads by 0.1 cycles the emission-line source.885 Lhe spin pulses in the line profiles are very clear in and show the same repetitive pattern (phase sult from red to blue velocities: see 7)., The spin pulses in the line profiles are very clear in and show the same repetitive pattern (phase shift from red to blue velocities; see 7).886 Trace of the red-to-blue. sinusoidal. component can also be discerned in the trailect profiles of 4471.," Trace of the red-to-blue, sinusoidal, component can also be discerned in the trailed profiles of 4471."887 We reconstruct images of the emission-line. distributions using the Doppler-shifted. line profiles. (railed spectra)., We reconstruct images of the emission-line distributions using the Doppler-shifted line profiles (trailed spectra).888 The Doppler image projected in a particular direction (orbital phase) is a line profile., The Doppler image projected in a particular direction (orbital phase) is a line profile.889 Phe back-projection imaging inverts the problem: ancl reconstructs the image from the ine profiles (Llorne 1991)., The back-projection imaging inverts the problem and reconstructs the image from the line profiles (Horne 1991).890 Examples ancl techniques of Doppler tomography (using either the linear back-projection or maximum entropy) have been successfully applied in cataclysmic variables anc X-ray binaries., Examples and techniques of Doppler tomography (using either the linear back-projection or maximum entropy) have been successfully applied in cataclysmic variables and X-ray binaries.891 The imagine echnique has resolved. emission. within the binary from ocations such as the red star (IP. Pee: Larlaltis οἱ al., The imaging technique has resolved emission within the binary from locations such as the red star (IP Peg; Harlaftis et al.892 1994). the gas stream (OY Car: Llarlaftis and. Marsh. 1996a: see here. inparticular. for linear back-projection image," 1994), the gas stream (OY Car; Harlaftis and Marsh 1996a; see here, inparticular, for linear back-projection image"893The origin of irregular satellites and (he mechanisms of their capture remain unknown.,The origin of irregular satellites and the mechanisms of their capture remain unknown.894 The high values of the orbital inclinations aud eccentricities of (hese objects imply an origin outside (he primordial circumplanetary disk from which the regular satellites of giant planets were formed., The high values of the orbital inclinations and eccentricities of these objects imply an origin outside the primordial circumplanetary disk from which the regular satellites of giant planets were formed.895 It is believed that irregular satellites were formed elsewhere and were captured in their current orbits (Ixuiper1956:Pollack.Burns.andTauber1979:Nesvorny2003:Nesvorn*.Deaugé.andDones2004:JewittIlaghighipour 2007).," It is believed that irregular satellites were formed elsewhere and were captured in their current orbits \citep{Kuiper56,Pollack79,Nesvorny03,Nesvorny04,Jewitt07}."896. The capture of inregular satellites might have occurred during aud/or after (he formation ol (he regular satellites of the giant planets., The capture of irregular satellites might have occurred during and/or after the formation of the regular satellites of the giant planets.897 Given that the latter objects are formed through the collisional growth: of small bodies in a circumplanetury disk (Canupand.Ward2002:AloscqueiraandEstrada2008a.b:Estrada:ancl\losqueira 2006).. the orbits of captured irreeulars might have been altered by perturbations [rom these objects during (heir formation id after they. are fully formed.," Given that the latter objects are formed through the collisional growth of small bodies in a circumplanetary disk \citep{Canup02,Mosqueira03a,Mosqueira03b,Estrada06}, the orbits of captured irregulars might have been altered by perturbations from these objects during their formation and after they are fully formed."898 In the case of Jovian irregulars. (he migrations of Ganvinecle id Callisto (TittemoreandWisdom1955.1989.1990:GoldreichTremaineidWard2002) have also had significant effects on the dvnamies of irregular satellites.," In the case of Jovian irregulars, the migrations of Ganymede and Callisto \citep{Tittemore88,Tittemore89,Tittemore90,Goldreich80,Canup02}899 have also had significant effects on the dynamics of irregular satellites."900 In this paper we study the dnamies and stability of irregular satellites between Callisto and Themisto., In this paper we study the dynamics and stability of irregular satellites between Callisto and Themisto.901 We present the details of our model in 32. and an analysis of the results in 83.," We present the details of our model in 2, and an analysis of the results in 3."902 Section 4 concludes this study by. reviewing our study and discussing its limitations., Section 4 concludes this study by reviewing our study and discussing its limitations.903 We numerically integrated the orbits of several hundred test particles in a region interior to the orbit of Themisto. the innermost Jovian irregular satellite.," We numerically integrated the orbits of several hundred test particles in a region interior to the orbit of Themisto, the innermost Jovian irregular satellite."904 We assumed that the regular satellites of Jupiter were fully formed. and. studied the perturbative effects of the Galilean satellites on the dvnaimics of small objects in their vicinities., We assumed that the regular satellites of Jupiter were fully formed and studied the perturbative effects of the Galilean satellites on the dynamics of small objects in their vicinities.905 We considered a svstem consisting of Jupiter. the Galilean satellites. and 500 lest particles unilormly distributed between 30 and 80 Jupiter-radii.," We considered a system consisting of Jupiter, the Galilean satellites, and 500 test particles uniformly distributed between 30 and 80 Jupiter-radii."906 The initial orbital elements of the test particles were chosen in asvstemalic wav as explained below., The initial orbital elements of the test particles were chosen in a systematic way as explained below.907" 1) At the beginning of each simulation. test particles were placed in orbits with semimajor axes starting al 20/2, and increasing in increments of 0.118."," 1) At the beginning of each simulation, test particles were placed in orbits with semimajor axes starting at $30{R_J}$ and increasing in increments of $R_J$."908" 2) For each initial value of the semimajor axis of a test particle (αρ). the initial orbital eccentricity (6,) was chosen to be 0. 0.2. 0.4. and 0.6."," 2) For each initial value of the semimajor axis of a test particle $(a_p)$, the initial orbital eccentricity $(e_p)$ was chosen to be 0, 0.2, 0.4, and 0.6."909 This choice of orbital eccentricity malches (he range of the curent. values of the orbital eccentricities of Jovian irregulars. as shown in figure 1.," This choice of orbital eccentricity matches the range of the current values of the orbital eccentricities of Jovian irregulars, as shown in figure 1."910and by the ddestruction reaction in the Le intershell. F((a.p)? Ne.,"and by the destruction reaction in the He intershell, $\alpha,p$ $^{22}$ Ne."911 In summary. the reactions (hat contribute to or affect the production of fuorine are: together with the alternative reaction chain The theoretical studies of Forestinietal.(1992) and found that the above described chain is activated in the convective pulse when neutrons are released. by {from the ILburning ashes.," In summary, the reactions that contribute to or affect the production of fluorine are: together with the alternative reaction chain The theoretical studies of \citet{forestini:92} and \citet*{mowlavi:96} found that the above described chain is activated in the convective pulse when neutrons are released by from the H-burning ashes."912 However. only the lowest observed abundances of ccould be explained.," However, only the lowest observed abundances of could be explained."913 An extra amount of iis required to produce the observed!°F.. and also to match the observed enhancements of eelements.," An extra amount of is required to produce the observed, and also to match the observed enhancements of elements."914 At the end of each TDU where the convective envelope expands into the stable radiative intershell zone extra-mixing processes could lead to the formation of a zone where protons and aare partially mixed (partial mixing zone)., At the end of each TDU where the convective envelope expands into the stable radiative intershell zone extra-mixing processes could lead to the formation of a zone where protons and are partially mixed (partial mixing zone).915 This would lead to additional production of bby the PC. 5) reaction in the top lavers of the Le intershell.," This would lead to additional production of by the $p,\gamma$ ) reaction in the top layers of the He intershell."916 Models including hvdrodynamical overshoot (lerwig2000).. rotation (Langeretal.1999). or the ellect. of gravity. waves (Denissenkov&Tout2003) have in fact produced a partial mixing zone resulting in the formation of apockel.. Stvanievoetal.(1995) showed that the {formed in the pocket is completely destroved bv the Οία. ΠΟ reaction before the onset of the next convective pulse.," Models including hydrodynamical overshoot \citep{herwig:00}, rotation \citep{langer:99} or the effect of gravity waves \citep{denissenkov:03} have in fact produced a partial mixing zone resulting in the formation of a. \citet{straniero:95} showed that the formed in the pocket is completely destroyed by the $\alpha,n$ $^{16}$ O reaction before the onset of the next convective pulse."917 Dy means of a parametric representation of the partial mixine zone Gallinoetal.(1998) and Goriely&\lowlavi(2000) showed (hat (his model can explain the observed properties of the iin AGB stars., By means of a parametric representation of the partial mixing zone \citet{gallino:98} and \citet{goriely:00} showed that this model can explain the observed properties of the in AGB stars.918 In the pocket |N is produced at conditions where the value of the proton to ratio is close to wnily (seealsoMowlavi.Jorissen&Arnould1998)., In the pocket $^{15}$ N is produced at conditions where the value of the proton to ratio is close to unity \citep*[see also][]{mowlavi:98}.919. This PN is converted into PF. when the pocket is ingested in the following convective pulse., This $^{15}$ N is converted into $^{19}$ F when the pocket is ingested in the following convective pulse.920 analvzed the effect of the presence of the partial mixing zone on the nucleosvntliesis ol fluorine., \citet{goriely:00} analyzed the effect of the presence of the partial mixing zone on the nucleosynthesis of fluorine.921 These authors concluded that also by taking into account the nucleosvnthesis in (he partial mixing zone only the less Inorine-enriched stars could be explained., These authors concluded that also by taking into account the nucleosynthesis in the partial mixing zone only the less fluorine-enriched stars could be explained.922 The possible, The possible923of interesting features.,of interesting features.924 In Fig. 5..," In Fig. \ref{2250spec},"925 we display a continuum-subtracted image of the eemission line from the two-dimensional FORSL spectrum., we display a continuum-subtracted image of the emission line from the two-dimensional FORS1 spectrum.926 Also cdisplavecl are the emission. line velocity shifts and FAWHIAL. evaluated on a pixel-by-pixel basis. with the spatial zero taken at the position of the continuum centroid.," Also displayed are the emission line velocity shifts and FWHM, evaluated on a pixel-by-pixel basis, with the spatial zero taken at the position of the continuum centroid."927 The EWHLIAI (after correcting for instrumental broadening) varies in the range ~150—410kms|. with a maximum at he position of the continuum. centroikd. (possibly due. to unresolved rotation in the central regions).," The FWHM (after correcting for instrumental broadening) varies in the range $\sim 150-410 \rm km\,s^{-1}$, with a maximum at the position of the continuum centroid (possibly due to unresolved rotation in the central regions)."928 The line emission is asynimetric at some spatial olfsets. where the line profile is xtter reproduced by multiple Caussian components.," The line emission is asymmetric at some spatial offsets, where the line profile is better reproduced by multiple Gaussian components."929 While he line broadening and asvmumetries might be expected o distort the velocity. profile. the fact that. it. remains remarkably smooth anc does not display sharp changes at he position of maximum line broadening suggests that the overall kinematic pattern of the gas is not severely distorted.," While the line broadening and asymmetries might be expected to distort the velocity profile, the fact that it remains remarkably smooth and does not display sharp changes at the position of maximum line broadening suggests that the overall kinematic pattern of the gas is not severely distorted."930 At this point. it is interesting to examine in detail the ugh surface brightness emission line features present in the UST imaging observations of “P05.," At this point, it is interesting to examine in detail the high surface brightness emission line features present in the HST imaging observations of T05."931 Figure 6/— displays a continuume-subtracted comission line image of the EEL.A with the contrast levels adjusted to best clisplay the morphology of the high surface rightness features in the central 3daresee.," Figure \ref{oiii_closeup} displays a continuum-subtracted emission line image of the EELR, with the contrast levels adjusted to best display the morphology of the high surface brightness features in the central $3-4$ arcsec."932 Line emitting material extends out from the galaxy nucleus along the same »osition angle as the spectroscopic slit. i.e. in the direction of the companion galaxy: the eastern. emission is brighter ancl has a less regular shape.," Line emitting material extends out from the galaxy nucleus along the same position angle as the spectroscopic slit, i.e. in the direction of the companion galaxy; the eastern emission is brighter and has a less regular shape."933 Painter line emission surrouncs hese structures. filling the spectroscopic slit.," Fainter line emission surrounds these structures, filling the spectroscopic slit."934 Lt is possible that the dillerence between the seeing 1.567) and the slit width (~1.37) could have introduced an erroneous instrumental profile: in the case of an unresolved emission feature which does not fill the slit completely. the instrumental profile would be 2.544 rrather than the 5.9+0.2A numeasured from the sky anc are lines.," It is possible that the difference between the seeing $\sim9350.56^{\prime\prime}$ ) and the slit width $\sim 1.3^{\prime\prime}$ ) could have introduced an erroneous instrumental profile: in the case of an unresolved emission feature which does not fill the slit completely, the instrumental profile would be $2.54$ rather than the $5.9 \pm 0.2$ measured from the sky and arc lines."936 Although the emission. line imaging observations confirm. that line emission. does indeed fll the FORSIL slit. (Fig. 6)).," Although the emission line imaging observations confirm that line emission does indeed fill the FORS1 slit (Fig. \ref{oiii_closeup}) ),"937 the brightest features clearly occupy a more confined. region within the hounds of the slit. and are therefore a potential problem.," the brightest features clearly occupy a more confined region within the bounds of the slit, and are therefore a potential problem."938 While this could potentially lead: το artificial velocity shifts. the close similarity between the kinematics derived. from the FORSL observations and the VIMOS observations suggest that there is no problem in this respect.," While this could potentially lead to artificial velocity shifts, the close similarity between the kinematics derived from the FORS1 observations and the VIMOS observations suggest that there is no problem in this respect."939 There are several possible interpretations for the observed. gas kinematics., There are several possible interpretations for the observed gas kinematics.940 On the basis of the FORSI spectrum alone. the gas kinematics are consistent with a  ldkpe diameter rotating disk. and an implied lower limit to the dvnamical mass of Asini|~210M. within a Ll aresec/d.55kpe radius.," On the basis of the FORS1 spectrum alone, the gas kinematics are consistent with a $\sim 15$ kpc diameter rotating disk, and an implied lower limit to the dynamical mass of $M sin i^{-1}941\sim 2 \times94210^{10} \rm M_{\odot}$ within a 1 arcsec/4.55kpc radius."943 Although this is a lower limit. the morphology of the ELLER. (rig. 6))," Although this is a lower limit, the morphology of the EELR (Fig. \ref{oiii_closeup}) )"944 would sugeest a close to eclec-on orientation for any disk system present., would suggest a close to edge-on orientation for any disk system present.945 Alternatively. the observed. velocities. could be caused by other means. with the observed. line. emission originating primarily from the edges of an ionization cone.," Alternatively, the observed velocities could be caused by other means, with the observed line emission originating primarily from the edges of an ionization cone."946 Close inspection of Fig., Close inspection of Fig.947 6— suggests that there is a dearth of emission.direcily alone the east-west aligned radio source axis. suggesting a hollowed-out region between he brightest emission. extending NE-SW ancl the fainter emission. extending SE-NW.," \ref{oiii_closeup} suggests that there is a dearth of emission along the east-west aligned radio source axis, suggesting a hollowed-out region between the brightest emission extending NE-SW and the fainter emission extending SE-NW."948 This is similar to the conic eatures observed in Cvgnus A Clavlor et al 2003: Tacdbiunter et al 1999: Jackson et al 1998)., This is similar to the conic features observed in Cygnus A (Taylor et al 2003; Tadhunter et al 1999; Jackson et al 1998).949 We note further that that he observed. velocity. amplitude clisplavecl in Fig., We note further that that the observed velocity amplitude displayed in Fig.950 5 is much less than that observed. in other galaxies with gas undergoing regular rotation (e.g. Tadhunter et al 1989. Daum et al 1990).," \ref{2250spec} is much less than that observed in other galaxies with gas undergoing regular rotation (e.g. Tadhunter et al 1989, Baum et al 1990)."951 Other alternative explanations which should. be considered. are. inflows or outllows imperlectly aligned with the radio source axis (see Fig. 6))., Other alternative explanations which should be considered are inflows or outflows imperfectly aligned with the radio source axis (see Fig. \ref{oiii_closeup}) ).952and Bolleretal. (1992))).,and \citet{b69}) ).953" The sample of AGN studied by Tuelleretal.(2008) is not limited by ray absorption, so it is useful to consider the range of Rip, spanned by theseAGN."," The sample of AGN studied by \citet{b37} is not limited by X-ray absorption, so it is useful to consider the range of $R_{ir/x}$ spanned by theseAGN."954 Fig., Fig.955" 10 is as Fig. 2,,"," \ref{fig:bat} is as Fig. \ref{fig:nojets},"956 but indicates those AGN from Tuelleretal.(2008)., but indicates those AGN from \citet{b37}.957". Once again, it is clear that this sample spans the existing ranges of R;,./; space and does not stake out otherwise unoccupied regions."," Once again, it is clear that this sample spans the existing ranges of $R_{ir/x}$ space and does not stake out otherwise unoccupied regions."958" In order to test the effect of AGN redshift on our sample, we studied an independent sample of 150 additional high-z AGN observed from deep surveys with XMM-Newton and Chandra (Miyajietal.2004;Polletta2006;Tajeretal. 2007)."," In order to test the effect of AGN redshift on our sample, we studied an independent sample of 150 additional high-z AGN observed from deep surveys with XMM-Newton and Chandra \citep{b28,b27,b32}."959". At the high redshifts of AGN in the deep surveys (typically z~ 1), an observed IR luminosity at 24um corresponds to an AGN-frame 12um IR luminosity (see e.g. Treister, Krolik Dullemond (2008)) and a 2-10 keV X-ray luminosity corresponds to an AGN-frame 4—20 keV X-ray luminosity (which will typically lie within a factor of a few of the actual 2-10keV luminosity), so luminosity ratios Rir/z for these deep AGN will systematically be a factor of a few higher on average than for the local AGN in Fig. 3.."," At the high redshifts of AGN in the deep surveys (typically $\sim 1$ ), an observed IR luminosity at $24\micron$ corresponds to an AGN-frame $12\micron$ IR luminosity (see e.g. Treister, Krolik Dullemond (2008)) and a 2-10 keV X-ray luminosity corresponds to an AGN-frame $4-20$ keV X-ray luminosity (which will typically lie within a factor of a few of the actual 2-10keV luminosity), so luminosity ratios $R_{ir/x}$ for these deep AGN will systematically be a factor of a few higher on average than for the local AGN in Fig. \ref{fig:irx12}."960" From Fig. 11,,"," From Fig. \ref{fig:deep},"961" even without considering the systematic overestimate of Rj-/z above, the high z AGN appear to closely follow the luminosity ratio for the Group 1 AGN in Fig."," even without considering the systematic overestimate of $R_{ir/x}$ above, the high z AGN appear to closely follow the luminosity ratio for the Group 1 AGN in Fig."962" 3 and very few lie in the Group 2 AGN region (8/150 have R;,/_> 100) or in the blank areas of our luminosity plot.", \ref{fig:irx12} and very few lie in the Group 2 AGN region (8/150 have $R_{ir/x}>100$ ) or in the blank areas of our luminosity plot.963" While there are clearly many caveats to using such deep survey data (not least assumptions about the AGN SED shape based on the IR spectrum), it is intruiging that, even with our very simple approach, the deep survey points should follow very closely the low-z Group 1 AGN dispersion and not fill blank regions of the luminosity ratio plots."," While there are clearly many caveats to using such deep survey data (not least assumptions about the AGN SED shape based on the IR spectrum), it is intruiging that, even with our very simple approach, the deep survey points should follow very closely the low-z Group 1 AGN dispersion and not fill blank regions of the luminosity ratio plots."964 It is also comforting that our main distribution of Group 1 AGN in Fig., It is also comforting that our main distribution of Group 1 AGN in Fig.965" 3 is also consistent with the dispersion of the population of AGN in Mushotzkyetal. (2008),, which, since the X-rays are 14-195keV, should be independent of X-ray absorption."," \ref{fig:irx12} is also consistent with the dispersion of the population of AGN in \citet{b71}, , which, since the X-rays are 14-195keV, should be independent of X-ray absorption."966surface density (the correspouding value for 33077 (aperture #11) is 0.0013).,surface density (the corresponding value for 3077 (aperture 1) is 0.0043).967 Within the uncertainties this value is close to the average dusttogas ratio of Moana /Mggp20.005 derived by. Draine et ((2007) for 12 spiral galaxies of the SINGS sample for which Spitzer and SCUBA measurements were available.," Within the uncertainties this value is close to the average dust–to–gas ratio of $_{\rm968dust}$ $_{\rm \hi}\approx0.005$ derived by Draine et (2007) for 12 spiral galaxies of the SINGS sample for which Spitzer and SCUBA measurements were available."969 As noted bv Draine et ((2007). a Algae ΑΙ ratio between 0.003 and 0.01 is consistent with ealactie moetallicities between 0.3 and 1 times solu. if a similar fraction of heavy elements is in the foxiu of dust as in the Milkv War.," As noted by Draine et (2007), a $_{\rm dust}$ $_{\rm \hi}$ ratio between 0.003 and 0.01 is consistent with galactic metallicities between 0.3 and 1 times solar, if a similar fraction of heavy elements is in the form of dust as in the Milky Way."970 Our derived dusttoeas ratio thus indicates that the tidal armi is substantially chemically euriched. in agreeimeut with the ieeion metallicities derived by Croxall et ((2009).," Our derived dust–to–gas ratio thus indicates that the tidal arm is substantially chemically enriched, in agreement with the region metallicities derived by Croxall et (2009)."971 For comparison. the Mua /Mgg ratio for metalpoor chwart regular galaxies in the(muore extended) M81 group of galaxies is up to one order of magnitude lower than found here(Walter et 22007).," For comparison, the $_{\rm dust}$ $_{\rm HI}$ ratio for metal–poor dwarf irregular galaxies in the (more extended) 81 group of galaxies is up to one order of magnitude lower than found here (Walter et 2007)."972 So far. we lave oulv considered the οσα»," So far, we have only considered the gas."973", Molecular gas has been mapped in the (ος0) transition iu the tidal feature over a restricted area (Walter et 11998. Teithausen Walter 2000. Walter et 22006)."," Molecular gas has been mapped in the CO(1–0) transition in the tidal feature over a restricted area (Walter et 1998, Heithausen Walter 2000, Walter et 2006)."974 The two main molecular complexes (regiou uumber 411 aud #22in Ueithausen Walter 2000) arc spatially coincident with aperture nunbers 10 and 11 in this study., The two main molecular complexes (region number 1 and 2 in Heithausen Walter 2000) are spatially coincident with aperture numbers 10 and 11 in this study.975 The implied molecular eas masses are dependent ou the choice of the Ho toCO conversion actor New., The implied molecular gas masses are dependent on the choice of the $_2$ –to–CO conversion factor $_{\rm CO}$.976 Ueithausen Walter (2000) used a conyersiou factor of Neog=8<lO2’em ?(K 1) |. a factor of ~L lueer than the Calactic conversiou actor.," Heithausen Walter (2000) used a conversion factor of $_{\rm977CO}$ $\times$ $^{20}$ $^{-2}$ $^{-1}$ $^{-1}$, a factor of $\sim$ 4 larger than the Galactic conversion factor."978 At the time. their choice was driven bv the asstuuption that the tidal region was mctalpoor.," At the time, their choice was driven by the assumption that the tidal region was metal–poor."979 Caven he abundance of dust aud dusttoσας ratios simular to jearby spiral galaxies. as well as the rregion uetallicitv Düeasurenienuts (indicatiug uctallicities as high as Galactic. Croxall et 22009). it low scClus to be more appropriate to use the Calactic conversion factor in this tidal feature.," Given the abundance of dust and dust–to–gas ratios similar to nearby spiral galaxies, as well as the region metallicity measurements (indicating metallicities as high as Galactic, Croxall et 2009), it now seems to be more appropriate to use the Galactic conversion factor in this tidal feature."980 This implies that he masses given in Heithauseu WalterMM(2000) should o divided bv a factor of Ll., This implies that the masses given in Heithausen Walter (2000) should be divided by a factor of 4.981 We thus molecular eas unasses of ed 4109 NIMES and 4109 MAL... for our apertures 10 and 11. respectively (comparable o the IT» mass preseut in the centro of 33077. Meier et 22001. Walter et 22002a).," We thus adopt molecular gas masses of $\sim$ $\times$ $^{6}$ $_\odot$ and $\sim$ $\times$ $^{6}$ $_\odot$ for our apertures 10 and 11, respectively (comparable to the $_2$ mass present in the centre of 3077, Meier et 2001, Walter et 2002a)."982 Averaged over our r—1kkpesized apertures this corresponds to a uolecular surface density between MAL. ppe? (i.c.. the ISM in the tidal feature is dominatedby the atomic gas phase even in the CObrightest regions.," Averaged over our kpc–sized apertures this corresponds to a molecular surface density between $_\odot$ $^{-2}$ (i.e., the ISM in the tidal feature is dominated by the atomic gas phase even in the CO–brightest regions."983 The detection of significaut amount of dus ο ο S106 in the tidal feature near 33077. distributed over 230. squarekiloparsecs. which is significantly larecr than the dust mass presen in the parent ealaxy (xrespoective of the choice of ο) aud reasonable temperatures). raises questions on the origin of the euriched material.," The detection of significant amount of dust $\sim$ $\pm$ $\times$ $^{6}$ $_\odot$ ) in the tidal feature near 3077, distributed over 30 square–kiloparsecs, which is significantly larger than the dust mass present in the parent galaxy (irrespective of the choice of $\beta$ and reasonable temperatures), raises questions on the origin of the enriched material."984 The ongoing star formation could potentially curich the medina in the tidal feature., The ongoing star formation could potentially enrich the medium in the tidal feature.985 If we asstuned a constant star formation rate since the creation of the feature (3.&105 vy ago. Yun ct 11991) we estimate a total mass of newly formed stars of ~ToLOMA... ie. less than the total zunount of dust that is preseut.," If we assumed a constant star formation rate since the creation of the feature $3\times10^8$ yr ago, Yun et 1994) we estimate a total mass of newly formed stars of $\sim7\times10^{5}$ $_{\odot}$, i.e. less than the total amount of dust that is present."986 This implies that the current rate of star formation activity can uot have created the prescut dust., This implies that the current rate of star formation activity can not have created the present dust.987 Also. the chemical eurichment due to the iregious in the tidal arm is expected to lead to a metallicitv of only Z~0.002 Solar (Walter ct 22006) over the last 3<LOS avr. Lie. inuch lower than what is ineasured by Croxall et ((2009) and implied by our dusttogas ratio.," Also, the chemical enrichment due to the regions in the tidal arm is expected to lead to a metallicity of only $\sim$ 0.002 Solar (Walter et 2006) over the last $3\times10^8$ yr, i.e., much lower than what is measured by Croxall et (2009) and implied by our dust–to–gas ratio."988 We conclude that the fidal armi nuatferial was preenriched. aud likely belonged to NGC32077 (which has a metallicity simular to the tidal iregious. Croxall et 22009) before the interaction.," We conclude that the tidal arm material was pre–enriched, and likely belonged to 3077 (which has a metallicity similar to the tidal regions, Croxall et 2009) before the interaction."989 Our fiudiugs imply that interactions between ealaxies can efficiently remove heavy clements. dust aud molecules from a galaxy.," Our findings imply that interactions between galaxies can efficiently remove heavy elements, dust and molecules from a galaxy."990 In the case discussed here. sjenificautlv more dust mass is found in the tidal ari than in the parent galaxy 33077 (tle siue holds true for the atomic ancl molecular eas).," In the case discussed here, significantly more dust mass is found in the tidal arm than in the parent galaxy 3077 (the same holds true for the atomic and molecular gas)."991 Interactions thus appear to have the potential to alter the chemical evolution of a galaxy dramatically aud to expel au iuterstellur medium that is enriched by heavy clemieuts., Interactions thus appear to have the potential to alter the chemical evolution of a galaxy dramatically and to expel an interstellar medium that is enriched by heavy elements.992 As interactions have been more frequent at larecr lookback times it is concervable that this mechanism can Gn addition to outflows. e.g.. Steidel et 22010) eHectively eurich the intergalactic medium (sce also Roussel et 22010. Walter ct 220025).," As interactions have been more frequent at larger lookback times it is conceivable that this mechanism can (in addition to outflows, e.g., Steidel et 2010) effectively enrich the intergalactic medium (see also Roussel et 2010, Walter et 2002b)."993 The tidal systena discussed. here would be classified as a damped Lymanalpha alxorber (DLA) with a cross section of roughly 30 kpe? 22008).(ee. Wolfe. Gawiser Prochaska 2005. Zwaan ct ," The tidal system discussed here would be classified as a damped Lyman–alpha absorber (DLA) with a cross section of roughly 30 $^2$ (e.g. Wolfe, Gawiser Prochaska 2005, Zwaan et 2008)."994FW acknowledges the hospitality of the Aspen Center or Physics., FW acknowledges the hospitality of the Aspen Center for Physics.995Herschel is au ESA space observatory with science instruneuts providedby European-led Principal Tuvestigator consortia aud with iniportaut participation roni NASA., is an ESA space observatory with science instruments provided by European-led Principal Investigator consortia and with important participation from NASA.996Figure 2. shows the continuum energy distribution of C/2007 N3 (Lulin).,Figure \ref{sed} shows the continuum energy distribution of C/2007 N3 (Lulin).997 In this figure observed magnitudes are plotted against the mean wavelength of the filter band., In this figure observed magnitudes are plotted against the mean wavelength of the filter band.998 For comparison. energy distribution of solar analogue star HD76151. which was observed during the observing run. is also plotted.," For comparison, energy distribution of solar analogue star HD76151, which was observed during the observing run, is also plotted."999 It is seen that during the observing run the comet colour is similar to that of the solar colour., It is seen that during the observing run the comet colour is similar to that of the solar colour.1000" On February 24 the comet is observed through two apertures: 26"" and 547 (projected diameter 7700 km and 16300 km respectively)."," On February 24 the comet is observed through two apertures: 26"" and 54"" (projected diameter 7700 km and 16300 km respectively)."1001 There is indication that comet colour through the larger aperture is slightly bluer., There is indication that comet colour through the larger aperture is slightly bluer.1002 This could be due to the disintegration of larger grains into smaller grains as they move out resulting in higher population of sub-micron size grains in the outer coma., This could be due to the disintegration of larger grains into smaller grains as they move out resulting in higher population of sub-micron size grains in the outer coma.1003" Observations on all the dates with 26"" aperture show comet colour similar to solar In the following we discuss the light curve and the phase curve of the comet."," Observations on all the dates with 26"" aperture show comet colour similar to solar In the following we discuss the light curve and the phase curve of the comet."1004 For this purpose. magnitudes. denoted as m1.1.0). are referred to at A and + equal to LAU by subtracting the term 2.5LogCGN +7) from the observed magnitudes.," For this purpose, magnitudes, denoted as $m(1,1,\alpha)$, are referred to at $\Delta$ and $r$ equal to 1AU by subtracting the term $\Delta ~r^2$ ) from the observed magnitudes."1005 The light curves (LCs ie. πα) vs time(JID)) of C/2007 N3 (Lulin) are plotted in Figure δέ) and the phase angles at the time of observation are marked., The light curves (LCs i.e. $\alpha$ ) vs time(JD)) of C/2007 N3 (Lulin) are plotted in Figure \ref{light_curve}( (a) and the phase angles at the time of observation are marked.1006 LCs in different filter bands are annotated in the figure., LCs in different filter bands are annotated in the figure.1007 It is seen that comet gets brighter as a decreases., It is seen that comet gets brighter as $\alpha$ decreases.1008 During the pre-opposition phase we have observations.only at two phase angles - 6.69 and 5.757. with the comet being brighter at lower phase angle.," During the pre-opposition phase we have observationsonly at two phase angles - 6.69 and $5.75^\circ$, with the comet being brighter at lower phase angle."1009 The phase curve is better covered during the post-opposition phase. the minimum phase angle. at which observations are made. being ~1.77.," The phase curve is better covered during the post-opposition phase, the minimum phase angle, at which observations are made, being $\sim 1.7^\circ$."1010 The comet was not observable when it was close to zero phase (local day time)., The comet was not observable when it was close to zero phase (local day time).1011 During post-opposition phase. a clear increase in brightness with decreasing @ is observed.," During post-opposition phase, a clear increase in brightness with decreasing $\alpha$ is observed."1012 This is consistent in all the filter bands., This is consistent in all the filter bands.1013 To look at the change in brightness with phase. we have plotted mtl.l1.a) vs à in Figure 3b).," To look at the change in brightness with phase, we have plotted $\alpha$ ) vs $\alpha$ in Figure \ref{light_curve}( (b)."1014 No opposition surge is noticed for the observed range in a. only some linear increase of brightness with decreasing à is detected.," No opposition surge is noticed for the observed range in $\alpha$, only some linear increase of brightness with decreasing $\alpha$ is detected."1015 We checked for the linearity in phase curve by plotting the flux against the phase angle and found it to be linear in all the bands., We checked for the linearity in phase curve by plotting the flux against the phase angle and found it to be linear in all the bands.1016 For discussion purpose in this paper we have adopted the phase curve (magnitude vs phase) re. Figure 3€)., For discussion purpose in this paper we have adopted the phase curve (magnitude vs phase) i.e. Figure \ref{light_curve}( (b).1017 Though we notice a linear brightness increase with decreasing a for 1.7<a 107. we do not have observations for à.«1.77 to comment on opposition surge (i.e.," Though we notice a linear brightness increase with decreasing $\alpha$ for $1.7^{\circ} <\alpha < 10^{\circ}$ , we do not have observations for $\alpha < 1.7^{\circ}$ to comment on opposition surge (i.e."1018"effect. bv comparing 5 with £,.","effect, by comparing $\gamma$ with $L_r$."1019 It also shows that. when the two classes are separated. there is no dependence of the radio-huninosity ou the value of 5. e.g. naller values of + in core galaxies are uot associated to the brightest racllo-SOurces.," It also shows that, when the two classes are separated, there is no dependence of the radio-luminosity on the value of $\gamma$, e.g. smaller values of $\gamma$ in core galaxies are not associated to the brightest radio-sources."1020 However. below this threshold. the two populations of carly-type galaxies cannot be readilv differentiated.," However, below this threshold, the two populations of early-type galaxies cannot be readily differentiated."1021 We are left with the ambienity on what is the driving imiechanismi at the origm of this threshold iu raclio-huninosity for power-law ealaxics: it can be due to an, We are left with the ambiguity on what is the driving mechanism at the origin of this threshold in radio-luminosity for power-law galaxies: it can be due to an1022The Durst. And Transient Source Experiment on the Compton Gamma Rav Observatory (Fishmanetal1989) provided near continuous monitoring of the whole sky at hard X-ray energies.,The Burst And Transient Source Experiment on the Compton Gamma Ray Observatory \cite{fish} provided near continuous monitoring of the whole sky at hard X-ray energies.1023 DATSE only sees the high energy ail of LIMINBs. typically deected in the lowest DISCLA channel and CONT channes l-4 covering 20-T0keV. In studying X-ray. pulsars two moces of operation are possible: »ulsed Dux and Earth occultaion.," BATSE only sees the high energy tail of HMXBs, typically detected in the lowest DISCLA channel and CONT channels 1-4 covering 20-70keV. In studying X-ray pulsars two modes of operation are possible; pulsed flux and Earth occultation."1024 HE pulsations are detected corresponding to the spin period of a known pulsar then he pulsed Lux and. pulse frecjuency can be measured. with Fourier analysis or epoch folcing (Dildsten. L. et al.," If pulsations are detected corresponding to the spin period of a known pulsar then the pulsed flux and pulse frequency can be measured with Fourier analysis or epoch folding (Bildsten, L. et al."1025 1997) and: used o monitor the source with great. sensitivity and ine time resolution., 1997) and used to monitor the source with great sensitivity and fine time resolution.1026 (Dildstenetal1997) For the long erm study of behavior related to orbital motion in binary systems. daily averages of the total lux are sullicient. and are in acleilion sensitive to non-pulsec X-ray output. from he source.," \cite{bil} For the long term study of behavior related to orbital motion in binary systems, daily averages of the total flux are sufficient and are in addition sensitive to non-pulsed X-ray output from the source."1027 By exploiting the Earth's imb as an occulting mask. individual X-rav sources can xf isolated with an accuracy of 17.," By exploiting the Earth's limb as an occulting mask, individual X-ray sources can be isolated with an accuracy of $\degr$."1028 Termed the Earth Occultation Technique. two measurements of the total ας from a source are mace during each 90 minute satellite «orbit as jo ποιος nioves into and out of eclipse due to the satellite’s motion. full details are given by Zhang et al (1994). llarmon et al (2002)..," Termed the Earth Occultation Technique, two measurements of the total flux from a source are made during each 90 minute satellite orbit as the source moves into and out of eclipse due to the satellite's motion, full details are given by Zhang et al \shortcite{zha} Harmon et al \shortcite{har}."1029 In this paper we present foldec X-ray dighteurves of Type E (normal) outbursts from S IIMXDs in which periodic Ductuations corresponding o previously. published orbital periods were significantIv. detected., 	In this paper we present folded X-ray lightcurves of Type I (normal) outbursts from 8 HMXBs in which periodic fluctuations corresponding to previously published orbital periods were significantly detected.1030 Such outbursts are thought to occur at cach periastron passage of the neutron star., Such outbursts are thought to occur at each periastron passage of the neutron star.1031 These outbursts last. for 8-10 days ancl the X-ray luminosity increases by a factor of 10., These outbursts last for 8-10 days and the X-ray luminosity increases by a factor of $\sim$ 10.1032 In contrast. Type LE are giant outbursts which do not correlate with the orbital phase but tend to last for several orbital eveles.," In contrast, Type II are giant outbursts which do not correlate with the orbital phase but tend to last for several orbital cycles."1033 The X-ray luminosity curing the outburst may reach close to the Eddington luminosity., The X-ray luminosity during the outburst may reach close to the Eddington luminosity.1034 1n adelition we studied the 30.5d modulation from LMC X-4 which was also a significantD cetection and investigated5 the N-rayv behaviour of A0538-668 in the light. of recent, In addition we studied the 30.5d modulation from LMC X-4 which was also a significant detection and investigated the X-ray behaviour of A0538-668 in the light of recent1035power spectra (Boughn&Crittenden2003) one could break the angular diameter distance degeneracy of the small scale anisotroples.,power spectra \citep{Boughn:2003yz} one could break the angular diameter distance degeneracy of the small scale anisotropies.1036 The interplay between perturbations in the dark energy and the IS\W is a subtle effect. whieh we will discuss in the section 27..., The interplay between perturbations in the dark energy and the ISW is a subtle effect which we will discuss in the section \ref{sec:gen}.1037 simple way to understand. the opposite behaviour of wo<l models is that for dw<1 the density in the dark energy component is. increasing with an expanding universe. while it is decreasing in a collapsing universe.," A simple way to understand the opposite behaviour of $w<-1$ models is that for $w<-1$ the density in the dark energy component is increasing with an expanding universe, while it is decreasing in a collapsing universe."1038 Hence the dark energy. perturbations are anfi-correlated with the matter perturbations as they are sourced., Hence the dark energy perturbations are correlated with the matter perturbations as they are sourced.1039 The bold lines in Fig., The bold lines in Fig.1040 1 correspond. to the case which includes perturbations., \ref{fig:ISW2} correspond to the case which includes perturbations.1041 Note that for «eπο1. the perturbations are exactly. zero.," Note that for $w=-1$, the perturbations are exactly zero."1042 We see how the bold. dot-dashed line (ee=/— 2) is significantly. lowered. compared to the thin line. due to the contribution of the perturbation pas. while for we=0.6 (dashed line) the contribution is significantly enhanced.," We see how the bold dot-dashed line $w=-2$ ) is significantly lowered compared to the thin line, due to the contribution of the perturbation $\delta\rho_{\rm de}$, while for $w=-0.6$ (dashed line) the contribution is significantly enhanced."1043 In Fig., In Fig.1044 3.2 we show the CMD temperature anisotropy spectrum for the three models. this time including »erturbations., \ref{fig:Clpert} we show the CMB temperature anisotropy spectrum for the three models this time including perturbations.1045 We clearly sec that the large —differences obtained on large scales when we οἷαnol include »erturbations in Fig., We clearly see that the large differences obtained on large scales when we did include perturbations in Fig.1046 2. have vanished., \ref{fig:Clno} have vanished.1047 This is because Or w>lo the smaller overall change in the background equation of state is enhanced. by the contribution due to he perturbations in the dark energy component., This is because for $w>-1$ the smaller overall change in the background equation of state is enhanced by the contribution due to the perturbations in the dark energy component.1048 For aw<1 the large contribution from the cillerent evolution of he background via the matter perturbations is partially cancelled by the contribution of the dark energy Iuctuation., For $w<-1$ the large contribution from the different evolution of the background via the matter perturbations is partially cancelled by the contribution of the dark energy fluctuation.1049 lt seems cdillieult to obtain information about the nature of dark energy from large scale CAIB information., It seems difficult to obtain information about the nature of dark energy from large scale CMB information.1050 We turn now to the problem of how to describe dark energy perturbations resolving to a scalar field., We turn now to the problem of how to describe dark energy perturbations resolving to a scalar field.1051" We should note as a reminder that we only resolved to a scalar field. in order to have a prescription for calculating the perturbations. where we assumed the most. simple kinetic term (0,47)ο"," We should note as a reminder that we only resolved to a scalar field in order to have a prescription for calculating the perturbations, where we assumed the most simple kinetic term $\pm (\partial_u1052\varphi)^2$."1053 These models have a speed. of sound ὃς2s=1., These models have a speed of sound $\hat{c}_s^2 = 1$.1054 llowever we have no idea what the dark energy. actually. is. so this assumption may be premature.," However we have no idea what the dark energy actually is, so this assumption may be premature."1055 For example. in a more generic class of dark cnerey models. so called. essence. the kinetic term does not need to be of such a simple form (Armencdariz-Piconctal.2000). and the sound speed generally differs from one.," For example, in a more generic class of dark energy models, so called k-essence, the kinetic term does not need to be of such a simple form \citep{Armendariz-Picon:2000dh} and the sound speed generally differs from one."1056 In the most general case the speed. of sound.end the equation of state evolve with timo. though clearly accounting for this is not feasible in general for parameter estimation.," In the most general case the speed of sound the equation of state evolve with time, though clearly accounting for this is not feasible in general for parameter estimation."1057 Llere we generalise the dark energy parameterisation by introducing a constant sound. speed 6 as a free parameter., Here we generalise the dark energy parameterisation by introducing a constant sound speed $\hcs$ as a free parameter.1058 If dq. ds initially zero. we see from Eqn.," If $\deltade$ is initially zero, we see from Eqn."1059 13. that it is sourced by the other perturbations iia=1 via the time evolution of the local scale factor. the source term 3(1| uM.," \ref{eqn:di1} that it is sourced by the other perturbations if $w\ne -1$ via the time evolution of the local scale factor, the source term $3(1+w)h'$ ."1060 An over density causes a decrease in the local expansion rate and so hb«0., An over density causes a decrease in the local expansion rate and so $h'<0$.1061 In this case a fluid starts to fall into overdensities i£;οαν but starts to fall out if uw;«1.," In this case a fluid starts to fall into overdensities if $w_i>-1$, but starts to fall out if $w_i<-1$."1062 The subsequent evolution depends on the sound. speed. as shown in Fig. 4..," The subsequent evolution depends on the sound speed, as shown in Fig. \ref{fig:kev}."1063 Consider the frame comoving with the dark matter (where 441= 0)., Consider the frame comoving with the dark matter (where $A=0$ ).1064 When &UH the term (1|wke; can be neglected. then the velocity and wavenumber only enter via the combination (1|«;)e;/&.," When $k\ll \H$ the term $(1+w_i)kv_i$ can be neglected, then the velocity and wavenumber only enter via the combination $(1+w_i)v_i/k$."1065" For large sound speeds the source term for the velocities is large anc they are anti-damped. which leads to an almost. A-independent evolution where the dark energy. perturbations change sign at carly times. and. become the sign to 0,,."," For large sound speeds the source term for the velocities is large and they are anti-damped, which leads to an almost $k$ -independent evolution where the dark energy perturbations change sign at early times, and become the sign to $\delta_m$."1066 At late times when the dark energv becomes a significant. fraction of the energy. density. the total density. perturbations are therefore smaller than without dark energy. perturbations. there is à larger overal change in the potential. and the ISNV contribution is increased.," At late times when the dark energy becomes a significant fraction of the energy density, the total density perturbations are therefore smaller than without dark energy perturbations, there is a larger overal change in the potential, and the ISW contribution is increased."1067 The signreversal happens later for lower sound speeds as we see in Fig., The signreversal happens later for lower sound speeds as we see in Fig.1068 4. and for ~18 the perturbations never reverse., \ref{fig:kev} and for $\hcs \sim 1/3$ the perturbations never reverse.1069 Thus the contribution to, Thus the contribution to1070"are calculated as Flux errors in io, ioo, i45 and ἴιας are dominated by photon shot noise while the theoretical error in polarization fraction was estimated performing error propagation through the previous equations.","are calculated as Flux errors in $i_{0}$ , $i_{90}$ , $i_{45}$ and $i_{135}$ are dominated by photon shot noise while the theoretical error in polarization fraction was estimated performing error propagation through the previous equations."1071" In addition, we calculated the errors in p using a Monte Carlo method, which returned values similar to those estimated from error propagation."," In addition, we calculated the errors in $p$ using a Monte Carlo method, which returned values similar to those estimated from error propagation."1072" The lo uncertainty in 0 was estimated (i) by applying the relation derived by Serkowski(1974) using standard error propagation, that is, Ty=287.65c,/p, when p/a25; or (ii) graphically with the aid of the curve proposed by Naghizadeh-Khouei&Clarke(1993) when p/o,«5."," The $\sigma$ uncertainty in $\theta$ was estimated ) by applying the relation derived by \citet{Serkowski74}1073 using standard error propagation, that is, $\sigma_{\theta} = 28\degr.65~\sigma_{p}/p$, when $p/\sigma_p1074\ge 5$; or ) graphically with the aid of the curve proposed by \citet{NKC93} when $p/\sigma_p < 5$."1075 Figure 3 shows the polarization uncertainty as a function of the J-band magnitude achieved with our LIRIS observations., Figure \ref{shot} shows the polarization uncertainty as a function of the $J$ -band magnitude achieved with our LIRIS observations.1076" The observed distribution suggests that the uncertainties are dominated by photon shot noise, as expected for a sample collected with fixed exposure time."," The observed distribution suggests that the uncertainties are dominated by photon shot noise, as expected for a sample collected with fixed exposure time."1077" As expected, the uncertainties decrease when the data taken at aand aare combined."," As expected, the uncertainties decrease when the data taken at and are combined."1078 There is a natural limit which is due to the uncertainty bias when measuring low levels of polarization., There is a natural limit which is due to the uncertainty bias when measuring low levels of polarization.1079" Bias in the degree of linear polarization (p) comes from the fact that this quantity is defined as a quadratic sum of q and u, which produces a non-zero polarization estimate due to the uncertainties in their measurement (foradetaileddiscussionseeforinstance,Simmons&Stewart1985;WardleKronberg 1974).."," Bias in the degree of linear polarization $p$ ) comes from the fact that this quantity is defined as a quadratic sum of $q$ and $u$, which produces a non-zero polarization estimate due to the uncertainties in their measurement \citep[for1080a detailed discussion see for1081instance,][]{Simmons85,Wardle74}. ."1082" In order to remove the polarization bias and compute the true polarization, we used theprescription proposedby Simmons&Stewart(1985) for low polarization stars."," In order to remove the polarization bias and compute the true polarization, we used theprescription proposedby \citet{Simmons85} for low polarization stars."1083 The true polarization degree can be approximated by theexpressions Pye=0 , The true polarization degree can be approximated by theexpressions $p_{true} = 0$ 1084the “compact” sample.,the 'compact' sample.1085 The non-zero mean value of ya obtained above may results from a somewhat large uncertainty of individual object., The non-zero mean value of $y_{\rm d}$ obtained above may results from a somewhat large uncertainty of individual object.1086 If we assume that the sample regions compose a torial population. we can test whether the mean jq of the »opulation is regarded to be zero.," If we assume that the sample regions compose a normal population, we can test whether the mean $y_{\rm d}$ of the population is regarded to be zero."1087 If the variance of vq for he population is estimated to be square of the uncertaiuty of individual value. (0.2)7. we cau reject the Lypotlesis of he mean 4=0 with the significance level of or the “compact” sample. aud for the “ultra-compact” sample.," If the variance of $y_{\rm d}$ for the population is estimated to be square of the uncertainty of individual value, $(0.2)^2$, we can reject the hypothesis of the mean $y_{\rm d}=0$ with the significance level of for the 'compact' sample, and for the 'ultra-compact' sample."1088 Therefore. even if we take account of somewhat large uncertainties of the observations aud the uodel. the ionizing photon fraction cannot be reproduced o» a nebula filled with[um dust uniformly (ne. 44=0).," Therefore, even if we take account of somewhat large uncertainties of the observations and the model, the ionizing photon fraction cannot be reproduced by a nebula filled with dust uniformly (i.e. $y_{\rm d}=0$ )."1089 The dust distribution with the ceutral cavity is supported by our analysis indepeudently of the IR SED fitting or the photometric profile analysis., The dust distribution with the central cavity is supported by our analysis independently of the IR SED fitting or the photometric profile analysis.1090 The cavitvs radius in units of the Stromuneren radius (ec. ya) of the ultra-compact! sample is larger than that of the “compact” sample.," The cavity's radius in units of the Strömmgren radius (i.e. $y_{\rm1091d}$ ) of the 'ultra-compact' sample is larger than that of the 'compact' sample."1092" The real scale of the cavitys radii of both samples are about 0.3 pc aud 0.2 pec for ‘compact? aud ultra-compact’. respectively,"," The real scale of the cavity's radii of both samples are about 0.3 pc and 0.2 pc for 'compact' and 'ultra-compact', respectively."1093 We av see an evolutionary sequence of the dust cavity from ultra-conrpact to ‘compact’., We may see an evolutionary sequence of the dust cavity from 'ultra-compact' to 'compact'.1094 However. these values are agree with cach other within their error bars. and the “ultra-compact” sample consists of ouly three regions.," However, these values are agree with each other within their error bars, and the 'ultra-compact' sample consists of only three regions."1095 Thus. it is uncertain whether the difference hetween the cavitw’s radii of two subsaniples is real or not.," Thus, it is uncertain whether the difference between the cavity's radii of two subsamples is real or not."1096 Finally. we demonstrate the effect of the change of Ry on the obtained dust cavitw’s radius. yq.," Finally, we demonstrate the effect of the change of $R_V$ on the obtained dust cavity's radius, $y_{\rm d}$."1097 As show in Figure 2 (a) the value of yq iucreascs for a fixed ioniziug photon fraction. f. if Ry decreases.," As shown in Figure 2 (a), the value of $y_{\rm d}$ increases for a fixed ionizing photon fraction, $f$, if $R_V$ decreases."1098 This is shown quantitatively for three sample set in Table L., This is shown quantitatively for three sample set in Table 4.1099 For the ‘compact suple. the value of yq of Ry=3.1 is a factor of 1.7 times larger than that of Ry=5.5.," For the 'compact' sample, the value of $y_{\rm d}$ of $R_V=3.1$ is a factor of 1.7 times larger than that of $R_V=5.5$."1100 As described above. assuming spherical svuuuectry aud uuiform deusitv distribution. we obtain a typical radius of the central dust cavity in the compact! IT regious. 0.3 times Strónuugreu radius.," As described above, assuming spherical symmetry and uniform density distribution, we obtain a typical radius of the central dust cavity in the 'compact' H regions, 0.3 times Strömmgren radius."1101 This corresponds about 10 of the ionized radius aud about 0.3 pc., This corresponds about 40 of the ionized radius and about 0.3 pc.1102 We discuss the formation mechanisin aud the detectability of the cavity., We discuss the formation mechanism and the detectability of the cavity.1103" However, we should keep iu müud that since real II regious. even compact or ultra-compact regions. show a conrplex structure. the radius of a real cavity may be different from 0.3 pe obtained by our model."," However, we should keep in mind that since real H regions, even compact or ultra-compact regions, show a complex structure, the radius of a real cavity may be different from 0.3 pc obtained by our model."1104 Three formation mechanisms of the ceutral dust cavity are naturally expected: (1) radiation pressure. (2) stellar wind by the ceutral source. aud (3) dust sublimation.," Three formation mechanisms of the central dust cavity are naturally expected: (1) radiation pressure, (2) stellar wind by the central source, and (3) dust sublimation."1105 Here we discuss whether the above mechanisms can produce the cavity and which is more effective., Here we discuss whether the above mechanisms can produce the cavity and which is more effective.1106 The effect of radiation pressure of the ceutral source ou the distribution of gas aud dust in IT regions is discussed iu detail by Gail&ουαντ(1979b)., The effect of radiation pressure of the central source on the distribution of gas and dust in H regions is discussed in detail by \cite{gai79b}.1107. According to thoi. the radiation force acting ou the dust erain produces the central cavity of dust aud eas (See their Fig.7).," According to them, the radiation force acting on the dust grain produces the central cavity of dust and gas (See their Fig.7)."1108 The radius of the cavity is about 20 of the ionized radius., The radius of the cavity is about 20 of the ionized radius.1109 Although we cannot directly compare their cavitys radius with that in this paper because of the ditfereuce between the adopted piuzüueter sets. the radiation pressure can produce a large cavity.," Although we cannot directly compare their cavity's radius with that in this paper because of the difference between the adopted parameter sets, the radiation pressure can produce a large cavity."1110 Also. stroug stellar wind of the massive stars mus contribute to produce the central cavity.," Also, strong stellar wind of the massive stars must contribute to produce the central cavity."1111 Indeed. the central low density cavity of σας is produced by the stellar wind (c.e.. Comerón1997)).," Indeed, the central low density cavity of gas is produced by the stellar wind (e.g., \citealt{com97}) )."1112 Since dust aud gas are wel coupled with each other (Cail&Sedliuavr1979b).. the gas cavity indicates the dust cavity.," Since dust and gas are well coupled with each other \citep{gai79b}, the gas cavity indicates the dust cavity."1113 Uufortunatelv. we caunot estimate quantitatively the contribution of the stellar wine to form the dust cavity here.," Unfortunately, we cannot estimate quantitatively the contribution of the stellar wind to form the dust cavity here."1114 It is bevoud the scope of this paper., It is beyond the scope of this paper.1115 We will resolve which of radiation pressure aix stellar wiud is dominant mechanisia to form the cavity iu our future work., We will resolve which of radiation pressure and stellar wind is dominant mechanism to form the cavity in our future work.1116 Tow about the effect of dust sublimation?, How about the effect of dust sublimation?1117 Iudeed. dust erains may subline in IHE regious because of the strong radiation from the central source.," Indeed, dust grains may sublime in H regions because of the strong radiation from the central source."1118 However. the radius of the dust cavity caused by the cust sublimation is about 10! pe (Mookerjoa&Ghosh1999).," However, the radius of the dust cavity caused by the dust sublimation is about $10^{-4}$ pc \citep{moo99}."1119. This is too sinall to produce the large cavity expected here., This is too small to produce the large cavity expected here.1120 Thus. the dust sublimation is not dominant mechanism to form the dust cavity.," Thus, the dust sublimation is not dominant mechanism to form the dust cavity."1121 Iun addition. we discuss the effect of the chuupv distribution of dust.," In addition, we discuss the effect of the clumpy distribution of dust."1122 Uuder the chuupy distribution. which is more realistic than the smooth one adopted here. we ay reproduce the ionizing photon fraction determined from observational data even if the dust clumps exist iu the central area of IT regious.," Under the clumpy distribution, which is more realistic than the smooth one adopted here, we may reproduce the ionizing photon fraction determined from observational data even if the dust clumps exist in the central area of H regions."1123 However we cousider that there is also the central cavity under the chumipy: distribution., However we consider that there is also the central cavity under the clumpy distribution.1124 This is because the radiation pressure and stellar wiud will blows out the dust clumps from the ceutral area;, This is because the radiation pressure and stellar wind will blows out the dust clumps from the central area.1125 We note that the expected radius of the cavity in chuupy medium mav become sualler than the determined oue i the previous section., We note that the expected radius of the cavity in clumpy medium may become smaller than the determined one in the previous section.1126 Such effect of the chuupiness should be clarified iu our future work., Such effect of the clumpiness should be clarified in our future work.1127 As stated by Wright(1973).. the best observational test of the dust cavity model is FIR (or subiillimeter) observatious with high angular resolution.," As stated by \cite{wri73}, the best observational test of the dust cavity model is FIR (or submillimeter) observations with high angular resolution."1128 Assuming a spherical and wnitorm structure. we expect that the dust cavity’s radius is about 0.3 pe for our sample of the Galactic compact II reeious.," Assuming a spherical and uniform structure, we expect that the dust cavity's radius is about 0.3 pc for our sample of the Galactic compact H regions."1129" It corresponds to about 60""/(D/cIpc). where D is the distance to the region from us."," It corresponds to about $60''/(D/{\rm kpc})$, where $D$ is the distance to the region from us."1130 Hore. we examine the detectability of such dust cavity of “compact or ultra-compact” IT regions by the preset and future IR facilities.," Here, we examine the detectability of such dust cavity of 'compact' or 'ultra-compact' H regions by the present and future IR facilities."1131 The future IR observational satellite of Japan. ASTRO- GRIS: Tutrared Tagine Surveyor) is plauned to lamuch iu 2001 by the Iustitute of Space aud Astronatical Scieuce of Japan.," The future IR observational satellite of Japan, ASTRO-F ; Infrared Imaging Surveyor) is planned to launch in 2004 by the Institute of Space and Astronautical Science of Japan."1132" It will offer modest angular resolutious of 30” 50"" at 50200 citepiuurS98..", It will offer modest angular resolutions of $30''$ $50''$ at 50–200 \\citep{mur98}.1133 We may detect the dust cavity of II reeious located within only about 1 kpc from us., We may detect the dust cavity of H regions located within only about 1 kpc from us.1134 However. ASTRO-F will survey the whole of the sky. fortunately.," However, ASTRO-F will survey the whole of the sky, fortunately."1135 Thus. we may detect may such closest cavities.," Thus, we may detect many such closest cavities."1136" SIRTF. which is the major upcomime IR satellite planned to lanuch in July 2002 by NASA. offers much higher angular resolutions of 2.5"" 16"" at 21160 If we observe the enüssion from dust beime thermally"," SIRTF, which is the major upcoming IR satellite planned to launch in July 2002 by NASA, offers much higher angular resolutions of $2.5''$ $16''$ at 24–160 If we observe the emission from dust being thermally"1137burning luminosities (blue dashed lines) get quite large (Ly.210L. in the initiating flash ab /= 0). Shen&Bildsten(2009) have shown that the heating from Πο burning always occurs on a Gmescale much longer than the local dvnamical time. an outcome confirmed in multi-dimensions by Mocáketal.(2008.22009).,"burning luminosities (blue dashed lines) get quite large $L_{\rm He}\approx 10^{9.3}L_\odot$ in the initiating flash at $t=0$ ), \citet{shen09} have shown that the heating from He burning always occurs on a timescale much longer than the local dynamical time, an outcome confirmed in multi-dimensions by \citet{mocak08,mocak09}."1138. The stellar response during the 2 Myr core flash phase is shown in Figuree 2. and form the basis for our asteroseismologicalex work., The stellar response during the 2 Myr core flash phase is shown in Figure \ref{fig:structure} and form the basis for our asteroseismological work.1139 The non-radial adiabatic mode structures for RGB and. red clump stars have been studied. with focus on the f=1 mixed modes that reveal core properties (Dziembowskietal.2011a.b:Jiangetal.2011:diMaio 2011).," The non-radial adiabatic mode structures for RGB and red clump stars have been studied, with focus on the $\ell=1$ mixed modes that reveal core properties \citep{dziembowski01, christen04, dupret09, montalban10,christen11,christen11kep, jiang11, dimauro11}."1140. These mixed modes have a p-mocde quality in the outer parts of the star (where (μον are excited by. convection. and ivpically have ny210 radial nodes). but penetrate into the stellar core as very. high order (ny> LOO) g-modes.," These mixed modes have a p-mode quality in the outer parts of the star (where they are excited by convection, and typically have $n_p\approx 10$ radial nodes), but penetrate into the stellar core as very high order $n_g>100$ ) g-modes."1141 This justifies our initial exploration in (he WIND limit (Unnoetal.οἱal.2010:Christensen-Dalsgaard 2011a).. where the local (at radius r) racial waveniumber. ft. is where w=2av is the mode frequency. e; is the sound speed. N7 isthe Brunt-Viiisiilla frequency. and 57=(C+1)/r? is the Lamb frequency.," This justifies our initial exploration in the WKB limit \citep{unno89,aerts10,christen11}, , where the local (at radius $r$ ) radial wavenumber, $k_r$, is where $\omega=2\pi \nu$ is the mode frequency, $c_s$ is the sound speed, $N^2$ isthe Brunt-Väiisälla frequency, and $S_\ell^2=c_s^2 \ell (\ell+1)/r^2$ is the Lamb frequency."1142 Figure 3. shows the propagation diagrams (radial profiles of .N. 54 and 55) lor Four phases of the M—1.07. moclel. all chosen when Av= με.," Figure \ref{fig:brunt} shows the propagation diagrams (radial profiles of $N$, $S_1$ and $S_2$ ) for four phases of the $M=1.0M_\odot$ model, all chosen when $\Delta \nu=4 \mu$ Hz."1143 These resulted in nearly the same values of µας& 28/lIz (denoted by the horizontal line)., These resulted in nearly the same values of $\nu_{\rm max}\approx 28 \mu$ Hz (denoted by the horizontal line).1144 From top to bottom. thepanels are for /=—60 Mis prior (o the initiating flash (on the RGB). /=1.4 Alves (after the initiating flash). /=1.6 Myrs (during a convective He burning subflash) and /=2.0 Mrs when the star is in the convective He core burning phase of (he red clump.," From top to bottom, thepanels are for $t=-60$ Myrs prior to the initiating flash (on the RGB), $t=1.4$ Myrs (after the initiating flash), $t=1.6$ Myrs (during a convective He burning subflash) and $t=2.0$ Myrs when the star is in the convective He core burning phase of the red clump."1145 Modes supported by the acoustic response of the envelope (i.e. p-modes) propagate in the outer parts of the star where «7>57 and «7>Αι, Modes supported by the acoustic response of the envelope (i.e. p-modes) propagate in the outer parts of the star where $\omega^2>S_\ell^2$ and $\omega^2>N^2$.1146 Since NV? is either zero or very small there. we set 242=47—57.," Since $N^2$ is either zero or very small there, we set $c_s^2 k_r^2=\omega^2-S_\ell^2$."1147 The eigenfrequencies are (hen found by setting fhyde2nyz., The eigenfrequencies are then found by setting $\int k_r dr \approx n_p \pi $.1148" In the extreme limit of iw?ssSs. (his simplifies to m,7=αἱ{dyος. where (he integral extends over the outer parts ofthe star where «7> 57."," In the extreme limit of $\omega^2 \gg S_\ell^2$, this simplifies to $n_p\pi=\omega \int dr/c_s$, where the integral extends over the outer parts ofthe star where $\omega^2>S_\ell^2$ ."1149" Tradition is to write this as p= n,Av. where the “large spacing” Av is defined with the integral over the whole star. a reasonably accurate representation."," Tradition is to write this as $\nu=n_p\Delta \nu$ , where the “large spacing” $\Delta \nu$ is defined with the integral over the whole star, a reasonably accurate representation."1150"were chosen, using the grain size distribution of the MRN model (?)..","were chosen, using the grain size distribution of the MRN model \citep{Mathis}."1151 A mixture of silicate and graphite was assumed., A mixture of silicate and graphite was assumed.1152" In addition to the original RADMC code, we introduced foreground extinction Ay, as otherwise all models failed to reproduce the deep silicate absorption feature."," In addition to the original RADMC code, we introduced foreground extinction $A_V$, as otherwise all models failed to reproduce the deep silicate absorption feature."1153 Parameters that were varied during the fitting process are shown in the following list together with their values for the best-fit RADMC model: This best-fit model is shown in Fig. 6.., Parameters that were varied during the fitting process are shown in the following list together with their values for the best-fit RADMC model: This best-fit model is shown in Fig. \ref{RADMCSpektrum}.1154" It provides a good fit to the visibilities as well as the SED up to 12 um. The RADMC model describes the central YSO and its circumstellar matter up to the outer disk radius of 50 AU, i.e. predicts the emission in the central 100 AU diameter region."," It provides a good fit to the visibilities as well as the SED up to $12\,\mu$ m. The RADMC model describes the central YSO and its circumstellar matter up to the outer disk radius of 50 AU, i.e. predicts the emission in the central 100 AU diameter region."1155" Since this area is completely inside the MIDI field-of-view, the model visibility can be directly compared to the observed visibilities."," Since this area is completely inside the MIDI field-of-view, the model visibility can be directly compared to the observed visibilities."1156" For the fits to the SED, however, we have to take into account that the very large beams of the far- observations must clearly include emission from the surrounding large-scale cloud, on spatial scales of 20000 AU (0.1 pc) and larger, which is far outside the r=50 AU model area."," For the fits to the SED, however, we have to take into account that the very large beams of the far-infrared observations must clearly include emission from the surrounding large-scale cloud, on spatial scales of $\sim 20\,000$ AU $\sim 0.1$ pc) and larger, which is far outside the $r=50$ AU model area."1157" This large-scale cloud emission should not be confused with a possible circumstellar envelope, but represents the molecular clump and the surrounding cloud in which IRS 1 is embedded (see discussion below)."," This large-scale cloud emission should not be confused with a possible circumstellar envelope, but represents the molecular clump and the surrounding cloud in which IRS 1 is embedded (see discussion below)."1158 The mm-maps presented by ? show that the size of this cloud is about 1’ (corresponding to ~55000 AU or ~0.27 pc).," The mm-maps presented by \citet{Schreyer1} show that the size of this cloud is about $1'$ (corresponding to $\sim 55\,000$ AU or $\sim 0.27$ pc)."1159" The effects of this large-scale cloud are as follows: our MIDI measurements are insensitive to this large-scale emission, because it is far more extended than the angular resolution (and also the field-of-view) of MIDI."," The effects of this large-scale cloud are as follows: our MIDI measurements are insensitive to this large-scale emission, because it is far more extended than the angular resolution (and also the field-of-view) of MIDI."1160" Since the large-scale cloud emission is completely over-resolved for MIDI, it will not affect the observed visibilities."," Since the large-scale cloud emission is completely over-resolved for MIDI, it will not affect the observed visibilities."1161" The observed fluxes, however, are expected to be strongly affected by the large-scale cloud."," The observed fluxes, however, are expected to be strongly affected by the large-scale cloud."1162" First, the cloud material in front of the embedded object IRS 1 will cause considerable extinction."," First, the cloud material in front of the embedded object IRS 1 will cause considerable extinction."1163 This effect is represented by the foreground extinction we use in our modeling., This effect is represented by the foreground extinction we use in our modeling.1164 The second effect is that the cool dust in the large-scale cloud will produce far-infrared and mm-emission., The second effect is that the cool dust in the large-scale cloud will produce far-infrared and mm-emission.1165" Owing to the large beam sizes of the far-infrared and mm-observations, a large fraction of the observed fluxes in these beams will result from the large-scale cloud, andnot from the central embedded object IRS 1."," Owing to the large beam sizes of the far-infrared and mm-observations, a large fraction of the observed fluxes in these beams will result from the large-scale cloud, andnot from the central embedded object IRS 1."1166 This explains why our RADMC model fluxes at wavelengths 220m are considerably lower than the observed fluxes.," This explains why our RADMC model fluxes at wavelengths $\ga 20\,\mu$ m are considerably lower than the observed fluxes."1167" To approximately include this large-scale cloud emission component in our model, we added to our SED model in Fig."," To approximately include this large-scale cloud emission component in our model, we added to our SED model in Fig."1168 6 two additional blackbody components., \ref{RADMCSpektrum} two additional blackbody components.1169" For the temperature of the first component, we used T=55 K, as determined by ? for the clump in which IRS 1 is embedded."," For the temperature of the first component, we used $T=55$ K, as determined by \citet{Schreyer1} for the clump in which IRS 1 is embedded."1170" For the second component, we used a temperature of 120 K, which is typical of warm dust around massive YSOs in so-called “hot-cores” (?).."," For the second component, we used a temperature of 120 K, which is typical of warm dust around massive YSOs in so-called “hot-cores” \citep{Herbst}."1171 Fig 6 shows that with the addition of these two components good agreement between the observed fluxes and the model can be achieved., Fig \ref{RADMCSpektrum} shows that with the addition of these two components good agreement between the observed fluxes and the model can be achieved.1172" An interesting question is whether this 120 K component could be considered as an envelope around the disk, as for one interpretation of hot cores as an infalling envelope undergoing an intense accretion phase (?).."," An interesting question is whether this 120 K component could be considered as an envelope around the disk, as for one interpretation of hot cores as an infalling envelope undergoing an intense accretion phase \citep{Osorio}. ."1173 The RADMC model of a disk without surrounding envelope provides a very good fit of both, The RADMC model of a disk without surrounding envelope provides a very good fit of both1174we construct to dilferent scalar fields over that plane.,we construct two different scalar fields over that plane.1175 The first corresponds to the field measured in the numerical runs. noted L.," The first corresponds to the field measured in the numerical runs, noted $L$."1176 Phe second is a predicted. version. noted. L/.," The second is a predicted version, noted $L^{\prime}$."1177 The values of £L'(0;.c;) are caleulated from. the neighboring points in L(a.e). as follows We construct now the following quantity where Nis the total number of points in the plane a- c.," The values of $L^{\prime}(\alpha_{i},\epsilon_{j})$ are calculated from the neighboring points in $L(\alpha,\epsilon)$, as follows We construct now the following quantity where $N$ is the total number of points in the plane $\alpha$ - $\epsilon$."1178 This quantity help us to define the predictability. with στ as the variance of the landscape The predictability is bounded to P?1., This quantity help us to define the predictability with $\sigma^{2}$ as the variance of the landscape The predictability is bounded to $P\leq 1$.1179 A value of P~ implies that the landscape is very smooth. while for values P?x0 he changesὃν from neighboring>e sites can be high.," A value of $P\sim 1$ implies that the landscape is very smooth, while for values $P\leq 0$ the changes from neighboring sites can be high."1180o We now turn to the results of the Fig.3.. where we plot the precictabilitv as a function of the logarithm of the mass ofthe host halo. for all the galaxies in our study.," We now turn to the results of the \ref{predictfigure}, where we plot the predictability as a function of the logarithm of the mass of the host halo, for all the galaxies in our study."1181 Starting with the total galaxy mass (stars and the gas) we can see that the galaxies have high predictabilitv. P0.9. in most of the cases.," Starting with the total galaxy mass (stars and the gas) we can see that the galaxies have high predictability, $P > 0.9$, in most of the cases."1182 The situation is quite dilferent for the stellar mass and the bolometric luminosity., The situation is quite different for the stellar mass and the bolometric luminosity.1183 La these cases the preclictability ranges almost evenly between 0<Px]. and we start seeing some fraction of points with negative wedictability.," In these cases the predictability ranges almost evenly between $0<P<1$, and we start seeing some fraction of points with negative predictability."1184" In the case of the (D.Y) colors and ο”. SDSS, magnitudes we are in a totally dillerent. ballpark as most of the landscapes have negative predietabilitv. with a ew points over the range 0«2I."," In the case of the $(B-V)$ colors and $_U$, $_r$ magnitudes we are in a totally different ballpark as most of the landscapes have negative predictability, with a few points over the range $0<P<1$."1185 The conclusion after these results is that we spot a andscape with a very predictabilitv £2?«0.9 we can be sure hat the galaxy is sitting in halo less massive than 310H Al.., The conclusion after these results is that we spot a landscape with a very predictability $P<0.9$ we can be sure that the galaxy is sitting in halo less massive than $3\times10^{11}$ $_\odot$.1186" In the same vein. when picking the central galaxy in a 1aloof mass >107 M.. surely the preclictability is going to » lower P«0.9. or negative in the case of (D1) colors and SDSS,. SDSS; magnitudes."," In the same vein, when picking the central galaxy in a haloof mass $> 10^{12}$ $_\odot$, surely the predictability is going to be lower $P<0.9$, or negative in the case of $(B-V)$ colors and $_r$, $_U$ magnitudes."1187 We explore now the second. wav of quantification of our results., We explore now the second way of quantification of our results.1188" Lt is based on the landscape variance σι, over the the 320 points in the a- plane (I2q.11)) and the predictability P (Eq.10)).", It is based on the landscape variance $\sigma_{\alpha\epsilon}$ over the the 320 points in the $\alpha$ $\epsilon$ plane \ref{land_variance}) ) and the predictability P \ref{predigo}) ).1189 We want to weight the andscape variance by the information obtained. through the predictabilitv £P., We want to weight the landscape variance by the information obtained through the predictability $P$.1190 Performing a normalization in this way we can have an idea about how much should be expected to vary a given galactic woperty after performinge a perturbation (da. dc).," Performing a normalization in this way we can have an idea about how much should be expected to vary a given galactic property after performing a perturbation $\delta \alpha, \delta\epsilon$ )."1191 Using the variance σι alone can be misleading in the case ofa high predietability landscape. because it could over-estimate the variation of performing a (δὰ. δε) perturbation.," Using the variance $\sigma_{\alpha\epsilon}$ alone can be misleading in the case of a high predictability landscape, because it could over-estimate the variation of performing a $\delta \alpha, \delta\epsilon$ ) perturbation."1192 We propose then. the P-weighted variance which has the property of being bound between 0 [or the possible values of the predictability oc Pel.," We propose then, the $P$ -weighted variance which has the property of being bound between $0\leq \sigma_{\alpha\epsilon1193 P}\leq\sigma_{\alpha\epsilon}$ for the possible values of the predictability $-\infty<P\leq 1$ ."1194 The general trend. (Fig.4)) shows a growth in the /?- predietabilitv with halo mass. consistent with the," The general trend \ref{weightedfigure}) ) shows a growth in the $P$ -weighted predictability with halo mass, consistent with the"1195Considerable progress is being made with integral field wut (IFPU) spectrographs on laree telescopes in unraveling the internal properties of star-forming ealaxies over 1l«iI,Considerable progress is being made with integral field unit (IFU) spectrographs on large telescopes in unraveling the internal properties of star-forming galaxies over $1<z<5$.1196 During this period. the nass assenablv of salaxies proceeds at a rapid pace and such measures promise valuable insight iuto the physical imechanisnis by which voune ealaxics grow.," During this period, the mass assembly of galaxies proceeds at a rapid pace and such measures promise valuable insight into the physical mechanisms by which young galaxies grow."1197 Early effort has focused on the dynamical properties (c.c. ForsterSchreiberetal.2009:LawoetJonesetal.2010:Cinerueci 2010)).," Early effort has focused on the dynamical properties (e.g. \citealt{Forster09, Law09, Jones10, Gnerucci10}) )."1198 These studies je revealed svstenis which are dispersion-domimnated with varving deerces of ordered rotation cousisteut witli sources Which may develop stable disks aud ceutral stellar lees., These studies have revealed systems which are dispersion-dominated with varying degrees of ordered rotation consistent with sources which may develop stable disks and central stellar bulges.1199 The turbulent motions may be associated with vigorous star-formation driveu by cold molecular gas acereted aloug filaments from the nearby. interealactic uediunu (Clacconietal.2010)., The turbulent motions may be associated with vigorous star-formation driven by cold molecular gas accreted along filaments from the nearby intergalactic medium \citep{Tacconi10}.1200. The highest resolution data. secured by coupling the angular magnification of strong eravitational leusiug with laser-assisted euide star adaptive optics (LOSAQO). has revealed the sizes of star-forming reegious in : 22-39 galaxies sugeesting that star formation ds primarily trigecred by eravitational instability rather than external merecrs (Starkctal.2008:Jonesetal.2010:Swiubanukct 2010).," The highest resolution data, secured by coupling the angular magnification of strong gravitational lensing with laser-assisted guide star adaptive optics (LGSAO), has revealed the sizes of star-forming regions in $z\simeq$ 2-3 galaxies suggesting that star formation is primarily triggered by gravitational instability rather than external mergers \citep{Stark08, Jones10, Swinbank10}."

Showing the first 1,200 of 10342 lines. Download the file for the rest.