A number of Main Sequence Dwarf stars with different spectral types were observed at Pa Beta and at Br Gamma. These stars are: Gl356 A (G8V), BS4458 (K0 V), BS5553 (K2V), Gl380 (K7V), Gl410 (M2V), Gl411 (M2V) and Gl406 (M6V). Their Pa Beta and Br Gamma spectra are shown in Figures 4.3 and 4.4 respectively.
From the two M2V stars observed at Br Gamma, Gl410 was the one chosen. The two stars have slightly different photospheric spectra, as can be seen from Figure 4.4. The Wallace & Hinkle (1996) atlas of cool stars in the K band also presents the spectrum of Gl411 which looks very similar to the spectrum presented here. These authors express their surprise to the fact that the spectrum of Gl411 is not as similar as expected to the spectrum of the K7V star they observed (61 Cyg B), given the similar spectral types of the two stars. The reason for the difference appears to be that Gl411 is deficient in metals (see Wallace & Hinkle 1996). Given that the K band spectrum of Gl410 presented here is, as expected, similar to that of the K7V star observed (Gl380) and also since the signal-to-noise ratio in its spectrum is slightly higher than that in Gl411's spectrum made one choose Gl410 as a M2V template for this wavelength region.
Figure 4.3: Pa Beta Template Stars
The template star to use in the veiling calculation was, with a very few exceptions discussed below, decided on the basis of the spectral type of the T Tauri star as indicated in the Herbig-Bell Catalogue [Herbig & Bell 1988]. Most stars in which photospheric lines were identified have spectral types between K7V and M2V (26 out of 38 stars with photospheric lines in the Pa Beta region). For these stars, either the K7V or the M2V templates were used. It should be noted that the veiling in the K7V template measured by the M2V template (Gl410) is zero.
For the four stars with later spectral type than M2V (up to M5V) the M2V template was used. This choice was made on the basis that the near infrared spectra of these stars (FN Tau, FP Tau, GH Tau and ZZ Tau) resembles more that of the M2V template than it does that of the M6V template, even in the case of FN Tau, a M5V star (compare spectra of FN Tau, FP Tau, GH Tau and ZZ Tau in Figures 3.12 and 3.13 with the spectra in Figure 4.3) and the use of the earlier spectral type seems appropriate.
Of the eight stars with spectral types earlier than K7V three are within two subclasses of K7V and lacking a better template Gl380 (K7V) was used, for DI Cep, a G8V star, Gl356 A, a G8V template star was used.
A K7V template was used in the remaining four T Tauri
stars, HP Tau, V773
Tau and GM Aur (K3V stars) and T Tau (K0V), giving the largest
mismatches between the spectral type attributed from optical spectroscopy
and the type of the template star used to compute the veiling. K7V was
adopted because on inspecting the Br Gamma
spectrum of all four stars, one
sees that the relative intensity of the two photospheric
lines near 2.179 
resemble more those of a K7V star than those of
K0 or even K2 dwarfs, where the Si line should be the stronger of the
two. Also, the photospheric lines seen near Pa Beta, in these T Tauri stars,
are deeper than the corresponding ones in the K0V and K2V template
stars, implying a later spectral type. On this basis, Gl380 (K7V) was
used as template. What is the reason for this discrepancy in spectral
types? The presence of redder unresolved (in CGS4) companions might be a
possibility. This is considered in Section 4.3 below. A different explanation is that
rather than seeing the stellar photosphere
one might be seeing the 'photosphere' of the circumstellar disk. In order
to obtain spectral types K7V to M2V the corresponding inner disk
temperatures must be of the order of 3500K. These sort of
temperatures might occur in accretion disks, however they would need to
extend to the stellar surface and accretion rates would have to be
high (
) [Calvet et al. 1997].
In principle, the four WTTS observed could have been used as template stars for the determination of veiling in the CTTS. One would expect their veiling to be zero, since they are non accreting T Tauri stars. Indeed, the veiling measurements for most of those stars (Hubble 4, LkCa 7 and DI Tau) are compatible with zero veiling, however, the values found are, with the exception of the K veiling of Hubble 4, not quite equal to zero (see Table 4.1). Therefore, rather than using these WTTS as templates, the main sequence stars observed were prefered. The veiling measurements on V773 Tau, a K3 V WTTS star, indicate high veiling for this star. It should be noted however that V773 Tau is a relatively fast rotator since its photospheric lines are quite broad, making its veiling determination harder. Due to this fact it was chosen not to use V773 Tau as a template star either.
Table 4.1 lists the optical spectral types of the T Tauri stars for which near infrared veiling was determined. The spectral types of the templates used to derive the near infrared veiling are also listed there.
A mismatch between the spectral type of the T Tauri star and that of the template star might lead to a wrong value for the veiling. In this work, this mismatch is seldom larger than two subclasses. This potential problem will be further discussed below.
