Datasets:
ReadingTimeMachine
/
rtm-sgt-ocr-v1

Dataset card Data Studio Files Community
2
source
stringlengths
1
2.05k
⌀
target
stringlengths
1
11.7k
The same pattern was carried out. 13 times in the Al-hane with 30. ουσοι ssec exposures. in total mmin.
The same pattern was carried out 13 times in the M-band with 30 coadded sec exposures, in total min.
These observations were obtained on the 3rd. of December. 2008. including a se of images for the bright AO star 1D36719. for calibration purposes.
These observations were obtained on the 3rd of December 2008, including a set of images for the bright A0 star HD36719 for calibration purposes.
All data was taken at low airmass (< 1.2): the airmass cilference between science and calibration targe was =O.1.
All data was taken at low airmass $<1.2$ ); the airmass difference between science and calibration target was $\la 0.1$.
For the reduction of the L- and Al-banel cata we followe the standard recipes based on the tools in the GeminisNIE package within ΕΛ.
For the reduction of the L- and M-band data we followed the standard recipes based on the tools in the Gemini/NIRI package within IRAF.
Flats and sky images were ereatec based. on the dithered. science images. discarding the firs frame.
Flats and sky images were created based on the dithered science images, discarding the first frame.
In the L-band all nine available images were usec for the sky frame.
In the L-band all nine available images were used for the sky frame.
The M-band sequence is long enough for the sky to change significantly.
The M-band sequence is long enough for the sky to change significantly.
For each individual frame we used the nine frames taken closest in time to create the sky image.
For each individual frame we used the nine frames taken closest in time to create the sky image.
After subtracting the sky the science frames were divided by the normalized Hatfield.
After subtracting the sky the science frames were divided by the normalized flatfield.
The resulting images were mostly [lat with the exception of the first and. last
The resulting images were mostly flat with the exception of the first and last
he other haud. this may be offset by having more uaterial to start with. and other features of the dead zone nav help provide “traps” where planetesimal can orm efiicieutly (Rice et al.
the other hand, this may be offset by having more material to start with, and other features of the dead zone may help provide “traps” where planetesimal can form efficiently (Rice et al.
2001) axd iueration cau be jalted or even reversed (see the following section).
2004) and migration can be halted or even reversed (see the following section).
lu this connection it is worth uoine the existence of he so-called “transition” disks. where outer. optically-hick disks surrounding T Tauri sars have very large inner disk holes - either partially or fully evacuated (e.g. Exspaillat 2007).
In this connection it is worth noting the existence of the so-called “transition” disks, where outer, optically-thick disks surrounding T Tauri stars have very large inner disk holes - either partially or fully evacuated (e.g, Espaillat 2007).
Tn some cases hese laree ΠΙΟ disk roles παν be the result of tidal toreues from companion (binary) stars. but in others eiaut planets may be the cause of the inner disk clearing.
In some cases these large inner disk holes may be the result of tidal torques from companion (binary) stars, but in others giant planets may be the cause of the inner disk clearing.
Iji the latter case. it nay be necessary to have multiple elaut planets form rearly simultaneously over a range of radii iu order to explain the sizes of these disk holes.
In the latter case, it may be necessary to have multiple giant planets form nearly simultaneously over a range of radii in order to explain the sizes of these disk holes.
The dead zones of our models. with relatively large surface deusities over a significant range of radii. may be able to promote the necessary runaway growth.
The dead zones of our models, with relatively large surface densities over a significant range of radii, may be able to promote the necessary runaway growth.
The sharp deusity jump at the dead zone outer radius (Ry) can significantly affect the planet migration outside the dead zoue CMatsunura Pudritz 2007): the associated torque on the planet depends on Bp. the density jump factor F. aud the width of the density jump.
The sharp density jump at the dead zone outer radius $_{D}$ ) can significantly affect the planet migration outside the dead zone (Matsumura Pudritz 2007); the associated torque on the planet depends on $_{D}$, the density jump factor $F$, and the width of the density jump.
Tu the framework of this paper. we derive tle first two parameters analytically aud test it uunericallv.
In the framework of this paper, we derive the first two parameters analytically and test it numerically.
For Bp. 83.1 and the Appeudix have shown that it starts at the outer disk at the eud of the iufall. aud then travels imwards with time at a speed given iu equation ALG iu the Appendix.
For $_{D}$, 3.1 and the Appendix have shown that it starts at the outer disk at the end of the infall, and then travels inwards with time at a speed given in equation \ref{eq:vd} in the Appendix.
With the initial Ro~R.oj.. Gf we oulv consider the slow rotating core: no GI fragmentation and active laver QI as discussed in the Appendix) and the speed of Rp. the position of Ry can be derived at ασ time later (Figure D11)).
With the initial $_{D}$$\sim$ $_{cmax}$ (if we only consider the slow rotating core: no GI fragmentation and active layer GI as discussed in the Appendix) and the speed of $_{D}$, the position of $_{D}$ can be derived at any time later (Figure \ref{fig:tran}) ).
Based on equation. A16.. the speed of Rp ects slower when moving inwards. which is seen iu Figure D11..
Based on equation \ref{eq:vd}, the speed of $_{D}$ gets slower when moving inwards, which is seen in Figure \ref{fig:tran}.
We can derive F if the dead zone residual α is simall enough (a<10
We can derive $F$ if the dead zone residual $\alpha$ is small enough $\alpha<10^{-5}$ ).
With such a small dead zone residual viscosity. the disk is eravitationally unstable withiu Ry.
With such a small dead zone residual viscosity, the disk is gravitationally unstable within $_{D}$.
Thus the surface density Jump cau be simply derived by dividing the surface density of the Q=1 disk to X4.. Here we have assumned the temperature profile as T=200 K(R/1 £2,
Thus the surface density jump can be simply derived by dividing the surface density of the Q=1 disk to $\Sigma_{A}$, Here we have assumed the temperature profile as $T$ =200 K(R/1 $^{-1/2}$.
Thus. as the dead zone moves inwards (Ap becomes αμα). Fo iuereases.
Thus, as the dead zone moves inwards $R_{D}$ becomes smaller), $F$ increases.
However. the density juup width. which is the last parameter required to calculate the torque on the planet. caunot be constrained by our LD simulation.
However, the density jump width, which is the last parameter required to calculate the torque on the planet, cannot be constrained by our 1D simulation.
Although at the early stage the inner disk is massive. at later stage (ολ) the outer disk (bevoud LO AU) is conrparable to the minima mass solar uchulae (MMSN) from Weideuschilliug (1977) (Fieure 1121).
Although at the early stage the inner disk is massive, at later stage $\sim$ Myr) the outer disk (beyond 10 AU) is comparable to the minimum mass solar nebulae (MMSN) from Weidenschilling (1977) (Figure \ref{fig:desch}) ).
Due to the boundary effect at the dead zone outer radius Rp (83.1 and Appendix). the outer disk evolves towards XxRLo as in the standard MMSN.
Due to the boundary effect at the dead zone outer radius $_{D}$ 3.1 and Appendix), the outer disk evolves towards $\Sigma\propto R^{-1.5}$ as in the standard MMSN.
Furthermore. if the dead zone Is mnmassive. Ry moves imvards verv slowly (81.5): then MGR1° lasts for a long time.
Furthermore, if the dead zone is massive, $_{D}$ moves inwards very slowly 4.5); then $\Sigma\propto R^{-1.5}$ lasts for a long time.
If plancts forma iu a massive dead zone. they may be lost by inward mueration: however. some may be trapped at the inner boundary (ANretke 2009) or outer boundary (Matsuninura 2007.2009) of the dead zone.
If planets form in a massive dead zone, they may be lost by inward migration; however, some may be trapped at the inner boundary (Kretke 2009) or outer boundary (Matsumura 2007,2009) of the dead zone.
Tn this paper. we have constructed a one-dimensional wo-zone accretion disk model to study disk formation and long-term evolution under the collapse of a BE rotating core.
In this paper, we have constructed a one-dimensional two-zone accretion disk model to study disk formation and long-term evolution under the collapse of a BE rotating core.
The model evolution can be divided into hree stages.
The model evolution can be divided into three stages.
At the carly stage. when the mass falls to he inner disk within AU scale. the MBI can be sustained in the immer disk aud efficiently aud steadily trausfers the iufalling mass to the central star.
At the early stage, when the mass falls to the inner disk within AU scale, the MRI can be sustained in the inner disk and efficiently and steadily transfers the infalling mass to the central star.
Later. when the mass ‘alls bevoud AU scale. the disk goes to the outburst stage due to the accretion rates’ uusmateh by the MBI aud GI as described iu Paper I. After the infall completes. the disk cuters the T Tauri phase and evolves on its own.
Later, when the mass falls beyond AU scale, the disk goes to the outburst stage due to the accretion rates' mismatch by the MRI and GI as described in Paper I. After the infall completes, the disk enters the T Tauri phase and evolves on its own.
Cores with ligher initial rotation cud up with a more massive disk aud more disk episodic accretion events (outbursts).
Cores with higher initial rotation end up with a more massive disk and more disk episodic accretion events (outbursts).
As long as the initial cloud core does not rotate extremely slowly to form a tiny disk (R41 AU). more than half of the star mass is built up by outbursts. which cases the "Iuninuositv problem.
As long as the initial cloud core does not rotate extremely slowly to form a tiny disk $_{cmax}$$\sim$ 1 AU), more than half of the star mass is built up by outbursts, which eases the “luminosity” problem.
Disks exhibit a variety of behavior during the T Tauri phase.
Disks exhibit a variety of behavior during the T Tauri phase.
For a disk with accretion sustained only by eravitational instability. the disk evolves towards a Q=1 disk aud the disk truucates at a radius sliehtlv larger than the maxinnun ceutrifueal radius of the iufall
For a disk with accretion sustained only by gravitational instability, the disk evolves towards a Q=1 disk and the disk truncates at a radius slightly larger than the maximum centrifugal radius of the infall.
If the disk has an active laver at the surface. however. the active laver can extend to a unich larger radius aud a sharp deusitv drop develops at a characteristic radius Ry that separates the mareinally eravitationally stable dead zone aud the MRI active but eravitationally stable outer disk.
If the disk has an active layer at the surface, however, the active layer can extend to a much larger radius and a sharp density drop develops at a characteristic radius $_{D}$ that separates the marginally gravitationally stable dead zone and the MRI active but gravitationally stable outer disk.
The density jump at Ry may be observable by the EVLA aud ALMA.
The density jump at $_{D}$ may be observable by the EVLA and ALMA.
The formation of a deuse belt of material is associated with the faihwe of magnetically dviven transport due to low ionization at intermediate radius in the disk: the only wavs to avoid this are (1) if there is a separate. equally efficient lvdrodvuamic rausport mechanisu. or (2) if for some reason the MBI zdls iu the outer disk as well perhaps due to dvuauuo zdlure.
The formation of a dense belt of material is associated with the failure of magnetically driven transport due to low ionization at intermediate radius in the disk; the only ways to avoid this are (1) if there is a separate, equally efficient hydrodynamic transport mechanism, or (2) if for some reason the MRI fails in the outer disk as well, perhaps due to dynamo failure.
We thank Sean M. Andrews aud David J. Wilner or kindly allowing us to use their figure iu our paper.
We thank Sean M. Andrews and David J. Wilner for kindly allowing us to use their figure in our paper.
ZZ thanks Robin Fowler for carefully reading the uanuscript aud imuproviug the writing.
ZZ thanks Robin Fowler for carefully reading the manuscript and improving the writing.
We thank the referee Cüuseppe Lodato for a helpful aud thorough report.
We thank the referee Giuseppe Lodato for a helpful and thorough report.
This work was supported iu part by NASA evant NNXOSAT39C€. bv the University of Michigan. by a Souv Faculty Fellowship. a Richard and Margaret Romano Professorial Scholarship. aud ai University Scholar appointinent to CC.
This work was supported in part by NASA grant NNX08A139G, by the University of Michigan, by a Sony Faculty Fellowship, a Richard and Margaret Romano Professorial Scholarship, and a University Scholar appointment to CG.
and the probability (hat no planets transit is For example. if (he planets are distributed isotropically then q(/)di=iundi. Qe=0 and gos(€41.62)=eyes.
and the probability that no planets transit is For example, if the planets are distributed isotropically then $q(i)di=\half\sin i\,di$, $Q_\ell=\delta_{\ell0}$ and $g_{22}(\epsilon_1,\epsilon_2)=\epsilon_1\epsilon_2$.
If the planets have zero inclination. it can be shown that although this result is derived more easily in other wavs.
If the planets have zero inclination, it can be shown that although this result is derived more easily in other ways.
These results can be extended to any number ofplanets’: where J, is the set of all permutations (py.....py) of the numbers L.....n. and m.xn.
These results can be extended to any number of: where $P_n$ is the set of all permutations $(p_1,\ldots,p_n)$ of the numbers $1,\ldots,n$, and $m\le n$ .
For example. The geometric selection matrix ήν refeq:gdefqe)) is simply (gn,UR/e4.P,fa2...Rfay.RR). the average of the geometric selection [actor over the joint distribution of stellar radius A, and planetary semi-major axis e [or the survev.
For example, The geometric selection matrix $G_{mn}$ \\ref{eq:gdefqq}) ) is simply $\langle g_{mn}(R_\star/a_1,R_\star/a_2,\ldots,R_\star/a_l,\bfkappa)\rangle$, the average of the geometric selection factor over the joint distribution of stellar radius $R_\star$ and planetary semi-major axis $a$ for the survey.
To evaluate G,,,(&) we use the separability assumption (4)) with respect (o e— ένα.Thus where [(ejdloge represents the probability distribution of e as modified by Che survey selectioneffects.
To evaluate $G_{mn}(\bfkappa)$ we use the separability assumption \ref{eq:sep}) ) with respect to $\epsilon=R_\star/a$ .Thus where $f(\epsilon)d\log\epsilon$ represents the probability distribution of $\epsilon$ as modified by the survey selectioneffects.
0.13 (Fig 20)).
0.13 (Fig \ref{fig:GaussFit}) ).
This time the fit successfully reproduces the asymmetry parameter distribution, including the low and high ends.
This time the fit successfully reproduces the asymmetry parameter distribution, including the low and high ends.
Only of the isolated galaxies are in excess of 3c.
Only of the isolated galaxies are in excess of $\sigma$.
The width of the half-Gaussian distribution sets an upper limit to the intrinsic dispersion of the aasymmetry in isolated galaxies.
The width of the half-Gaussian distribution sets an upper limit to the intrinsic dispersion of the asymmetry in isolated galaxies.
Errors in the calculation of the asymmetry index might be typically ~ 0.03 (mean of the Gaussian fit) (Sect. 3.2.3)).
Errors in the calculation of the asymmetry index might be typically $\sim$ 0.03 (mean of the Gaussian fit) (Sect. \ref{subsec:presentation}) ).
As discussed in Sect. 3.2.2},
As discussed in Sect. \ref{subsec:otheruncertainties},
there might also be random errors in the pointing (~ 0.04) and baseline subtraction (σ ~ 0.02) that may increase errors in Afluxratio.
there might also be random errors in the pointing $\sim$ 0.04) and baseline subtraction $\sigma$ $\sim$ 0.02) that may increase errors in $A_{flux~ratio}$.
Hence it is reasonable to expect a lower value of the width, c ~ 0.11, once these sources of errors are corrected.
Hence it is reasonable to expect a lower value of the width, $\sigma$ $\sim$ 0.11, once these sources of errors are corrected.
Note that the quantification of the asymmetry distribution for the galaxies in the rrefined subsample is not affected by inclination effects (e.g.?)..
Note that the quantification of the asymmetry distribution for the galaxies in the refined subsample is not affected by inclination effects \citep[e.g.][]{2009PhR...471...75J}.
Figure Ipshow sthattheinclinationo f thegalaxie saredistributedhéhlogeneouslyabovei 30°.
Figure \ref{fig:inclination}$ $a$ shows that the inclination of the galaxies are distributed homogeneously above $i$ = $\rm ^o$.
Only two galaxies have an inclination i < 15° (CIG 85 and 178).
Only two galaxies have an inclination $i$ $<$ $\arcdeg$ (CIG 85 and 178).
The lack of galaxies below i = 30 is caused by the width-to-channel ratio criterion explained in Sect. D.T].
The lack of galaxies below $i$ = 30 is caused by the width-to-channel ratio criterion explained in Sect. \ref{sub:sampleselection}.
This homogeneity in the inclination ensures that most of our galaxies do not show symmetric profiles because the galaxies are viewed face-on, where an asymmetry in the velocity field would remain unnoticed.
This homogeneity in the inclination ensures that most of our galaxies do not show symmetric profiles because the galaxies are viewed face-on, where an asymmetry in the velocity field would remain unnoticed.
To further inspect whether inclination can be introducing any bias in our results, we plotted the asymmetry index versus the inclination (Figure 2Ip)and. foundthatthetwoquantitiesarenotcorrelated.
To further inspect whether inclination can be introducing any bias in our results, we plotted the asymmetry index versus the inclination (Figure \ref{fig:inclination}$ $b$ ) and found that the two quantities are not correlated.
A reevaluation and quantification of isolation degree for CIG galaxies was reported in ??..
A reevaluation and quantification of isolation degree for CIG galaxies was reported in \citet{2007A&A...470..505V,2007A&A...472..121V}.
? derived two isolation parameters for each CIG galaxy: 1) a local surface density parameter 7x within the distance to the k-th neighbor (a good tracer of average galaxy surface density) and 2) a tidal strength parameter Q (a parameter more sensitive to interactions).
\citet{2007A&A...470..505V} derived two isolation parameters for each CIG galaxy: 1) a local surface density parameter $\eta_K$ within the distance to the k-th neighbor (a good tracer of average galaxy surface density) and 2) a tidal strength parameter $Q$ (a parameter more sensitive to one-on-one interactions).
iis known to be a sensitive diagnostic of interaction motivating us to compare these two parameters with our Apriratio asymmetry parameter.
is known to be a sensitive diagnostic of interaction motivating us to compare these two parameters with our $A_{flux~ratio}$ asymmetry parameter.
shows the lack of correlation between Αιratio and Figureboth 22]nx and Q.
Figure \ref{fig9} shows the lack of correlation between $A_{flux~ratio}$ and both $\eta_K$ and $Q$.
The Pearson's correlation coefficient is p = -0.005 and 0.114, respectively, which indicates that the two quantities are essentially not correlated.
The Pearson's correlation coefficient is $\rho$ = -0.005 and 0.114, respectively, which indicates that the two quantities are essentially not correlated.
A small trend in the Q parameter might be present in the sense that larger aasymmetries seem to be found in less isolated systems.
A small trend in the $Q$ parameter might be present in the sense that larger asymmetries seem to be found in less isolated systems.
The calculated intercept and slope are -3.3 +0.7 and 0.8 + 0.7, respectively.
The calculated intercept and slope are -3.3 $\pm$ 0.7 and 0.8 $\pm$ 0.7, respectively.
The lack of correlation suggest that we are minimizing nurture effects that might affect the sshape.
The lack of correlation suggest that we are minimizing nurture effects that might affect the shape.
The low values and small range in terms of galaxy density and tidal strength covered by CIG galaxies are not enough to see a correlation.
The low values and small range in terms of galaxy density and tidal strength covered by CIG galaxies are not enough to see a correlation.
? also suggest that there is no correlation between lopsidedness and tidal strength.
\citet{2005A&A...438..507B} also suggest that there is no correlation between lopsidedness and tidal strength.
However, they use the lopsidedness Αι parameter on NIR surface density, and NIR emission is not as extended as theI.
However, they use the lopsidedness $A_1$ parameter on NIR surface density, and NIR emission is not as extended as the.
We compare the asymmetry distribution of our rrefined subsample with that of different studies from the bibliography where a similar asymmetry index (Sect. 3.2))
We compare the asymmetry distribution of our refined subsample with that of different studies from the bibliography where a similar asymmetry index (Sect. \ref{sec:asymmetry_coefficients}) )
has been calculated and involve field/isolated galaxies (see also Sect.[I)) : ?,,
has been calculated and involve field/isolated galaxies (see also Sect. \ref{sec:introduction}) ): \citet{1998AJ....115...62H},
? and ?..
\citet{1998AJ....116.1169M} and \citet{2005A&A...438..507B}.
Figure D3| shows the Ayj,rario normalized distribution for our refined subsample with 1) ? and 2) a combined sample (N = 186) including ddata in ?,, ?,, and ? excluding CIG galaxies (80 galaxies).
Figure \ref{fig:comparison_hg98-2} shows the $A_{flux~ratio}$ normalized distribution for our refined subsample with 1) \citet{2005A&A...438..507B} and 2) a combined sample (N = 186) including data in \citet{1998AJ....116.1169M}, \citet{2005A&A...438..507B}, and \citet{1998AJ....115...62H} excluding CIG galaxies (80 galaxies).
The rrefined subsample shows the distribution best described by a half-Gaussian.
The refined subsample shows the distribution best described by a half-Gaussian.
It also shows the lowest absolute value of σ.
It also shows the lowest absolute value of $\sigma$.
The ? distribution shows the widest distribution (σ = 0.23) and noticeably deviates from a half-Gaussian curve.
The \citet{2005A&A...438..507B} distribution shows the widest distribution $\sigma$ = 0.23) and noticeably deviates from a half-Gaussian curve.
An intermediate case, σ = 0.17, is found for the combined sample without CIG galaxies.
An intermediate case, $\sigma$ = 0.17, is found for the combined sample without CIG galaxies.
Table [3] gives o values for each distribution as well as an asymmetry rate with "asymmetric" profiles defined as Afjyxratio values exceeding the 2σ level of our rrefined subsample (Arj,ratio = 1.26).
Table \ref{tab:afluxratio} gives $\sigma$ values for each distribution as well as an asymmetry rate with ”asymmetric” profiles defined as $A_{flux~ratio}$ values exceeding the $\sigma$ level of our refined subsample $A_{flux~ratio}$ = 1.26).
Figure compares the Afjyxratio cumulative probability distribution for our rrefined sample and those of ?,, ?,, and ?..
Figure \ref{fig:comparison_hg98} compares the $A_{flux~ratio}$ cumulative probability distribution for our refined sample and those of \citet{1998AJ....115...62H}, \citet{1998AJ....116.1169M}, and \citet{2005A&A...438..507B}.
In each plot the difference of the two curves indicates the asymmetry rate difference for a given Ayj;ratio limit.
In each plot the difference of the two curves indicates the asymmetry rate difference for a given $A_{flux~ratio}$ limit.
Our sample lies below the field samples in almost every bin with differences typically between 10 —20%.
Our sample lies below the field samples in almost every bin with differences typically between 10 –.
. A result more similar to our sample is found for ? likely in part because of the significant fractionof CIG galaxies (23%)) included in their sample.
A result more similar to our sample is found for \citet{1998AJ....115...62H} likely in part because of the significant fractionof CIG galaxies ) included in their sample.
Removing the CIG overlap increases both their σ and asymmetry rate.
Removing the CIG overlap increases both their $\sigma$ and asymmetry rate.
We performed a xy? test to check whether the null hypothesis that any of the three Afj.-ario distributions is similar to our rrefined sample, could be rejected.
We performed a $\chi^2$ test to check whether the null hypothesis that any of the three $A_{flux~ratio}$ distributions is similar to our refined sample, could be rejected.
Except for ? (y? = 9 and the associated p-value = 0.33) this hypothesis can be rejected.
Except for \citet{1998AJ....115...62H} $\chi^2$ = 9 and the associated p-value = 0.33) this hypothesis can be rejected.
In the cases of ? and 27 = 47 (p-value = 2 x 1077) and Y = 14 (p-value = 0.09) respectively.
In the cases of \citet{2005A&A...438..507B} and \citet{1998AJ....116.1169M}, $\chi^2$ $=$ 47 (p-value $=$ 2 $\times$ $^{-7}$ ) and $\chi^2$ $=$ 14 (p-value $=$ 0.09) respectively.
The sample differences we find are significant and cannot be ascribed to the refinement of the ssample (Sect. B.4)).
The sample differences we find are significant and cannot be ascribed to the refinement of the sample (Sect. \ref{sec:cleanedsample}) ).
In principle we do not know how much ? and ?’’s observations are affected by systematic errors, but differences involving the same criterion as used in our study would yield an asymmetry rate difference <
In principle we do not know how much \citet{1998AJ....116.1169M} and \citet{2005A&A...438..507B}' 's observations are affected by systematic errors, but differences involving the same criterion as used in our study would yield an asymmetry rate difference $<$ .
5%.. ? included only high S/N profiles, avoided pointing problems and quantified baseline problems, suggesting it is reasonable to compare it directly with our rrefined sample.
\citet{1998AJ....115...62H} included only high S/N profiles, avoided pointing problems and quantified baseline problems, suggesting it is reasonable to compare it directly with our refined sample.
Overall, because of their degree of isolation, our rrefined subsample and ?"s show a lower frequency ( 10
Overall, because of their degree of isolation, our refined subsample and \citeauthor{1998AJ....115...62H}' 's show a lower frequency $\sim$ 10
observations.
observations.
Use was made of the WEBDA database (http://www.univie.ac.at/webda). developed by J.-C.Mermilliod (Laboratory of Astrophysics of the EPFL. Switzerland) and operated by E. Paunzen al the University of Vienna. Austria.
Use was made of the WEBDA database (http://www.univie.ac.at/webda), developed by J.-C.Mermilliod (Laboratory of Astrophysics of the EPFL, Switzerland) and operated by E. Paunzen at the University of Vienna, Austria.
The author is very grateful to Werner W. Weiss for his constructive comments on the craft of this paper aud hiis scientific and organizational efforts. which made this work possible.
The author is very grateful to Werner W. Weiss for his constructive comments on the draft of this paper and his scientific and organizational efforts, which made this work possible.
She also acknowledges funding through the Austrian science Funds (FWE projects P-17580-N02 and T335-N16).
She also acknowledges funding through the Austrian Science Funds (FWF projects P-17580-N02 and T335-N16).