After bias and dark current subtraction, cosmic ray removal and flat field
correction, we used, for frame preprocessing, a software package written at
C.N.R. observing station at Monte Porzio (RM). Gradients of background
luminosity, due to imperfect flat field renormalization of non uniform sky
brightness, were eliminated by subtracting the least square plane of the
sky background image. The sky background in each band was estimated as the
average of the median values of 15 aperture photometry measurements using
(
) pxl. boxes located far from the galaxy. After this, all images
were convoluted twice with a two pixels FWHM gaussian filter. The
coordinates of each galaxy centre were calculated with a 0.2 pxl resolution,
fitting a gaussian profile with the FWHM of the nucleus. As all galaxies,
except NGC 6946, have small inclination angles, the radial surface
brightness was derived by integration over circular rings centered on the
galaxy. The luminosity of each ring, 1 pixel wide, was evaluated as
follows: around the average brightness value it was considered an
interval 10% of the average value wide. The surface brightness of the ring
was the median of the pixels set in this interval (for more details, see
Natali et al. 1992). For NGC 6946 some modifications were needed, to take
into account the significative inclination angle.
To derive the radial surface brightness for NGC 6946, we performed
integration over elliptical rings, which are circular in the plane of the
galaxy.
The position of the rings was specified by the inclination and position angles,
while the ring width and the luminosity calculations were the same as before.
We obtained the radial surface brightness profile in the five photometric
bands U, B, V, R, I, and the radial colour indexes (B-V), (B-I),
(U-B);
we also calculated
the radial scalelengths for the bulge and the disk. The scalelengths are
defined in different ways by different authors, namely
De Vaucouleurs (1974),
Freeman (1970), Boroson (1981),
Vila-Costas and Edmunds (1992), Kormendy (1977).
In this work we have adopted the definitions of Vila-Costas and
Edmunds for the bulge and of Freeman and Kormendy for the disk, so the
equations are, for the disk:
and for the bulge:
where:
&# m(r_e) & is the bulge half central magnitude m(r_d) & is 1/e the disk central magnitude m(0) & is the disk central magnitude r_e & is the bulge scalelength r_d & is the disk scalelength , & are the inverses of disk scalelengths.
We write the definition of the disk to bulge ratio as:
where:
&# S_d & is 1/e the disk central surface brightness S_e & is 1/e the bulge central surface brightness.
To obtain the disk and bulge scalelengths and the D/B parameter, we
fitted the disk using Eq. (1) or Eq. (2), obtaining 
; we subtracted the
disk profile to the total one and, fitting the bulge with Eq. (3), we
obtained 
. Finally we calculated D/B. The colour indexes were
calculated by subtracting each other the relative magnitude values for each
integration ring. In the following, discussion of the results obtained for
each galaxy from data analysis is reported.
truecm
NGC 1232:
truecm
This galaxy has been classified as peculiar by H. Arp, because of the
presence of a companion; its surface brightness radial profiles are
presented in Fig. 1 (click here).
Figure 1: Photometric profiles in UBVRI bands for all galaxies.
The magnitudes
have been rescaled by the indicated quantities for clarity's
sake
In Table 2 (click here) the values for the disk scalelengths for the five photometric bands, obtained using Eq. (1) are reported; in the second row there are the bulge scalelengths and in the third row the calculated values for the D/B parameter. As can be seen from the table, the disk scalelengths decrease toward longer wavelengths; perhaps there is an increase in scalelength for each new stellar population (RD).
There is a good correspondence between our D/B values and galaxy classification according van der Bergh, who assigns the subtype b to NGC 1232 (van der Bergh 1976). The colour indexes are given in Figs. 2 (click here)a and 2b. For (B-I) and (B-V) there is a pronounced decreasing trend toward larger radii, with a shallower slope between 10 and 20 Kpc.
Figure 2: a)
Radial trend of the colour indexes (B-V), (B-I) with
galactocentric radius
It is plausible that the interaction with its companion has changed the matter distribution throughout the galaxy, determining the shallow trend in the colour indexes. Between 20 and 25 Kpc there is a drop both in (B-I) and in (B-V), while the (U-B) index shows an opposite trend, see Fig. 2b.
Figure 2: b) Radial trend of the colour index (U-B) with
galactocentric radius
NGC 3184:
The radial surface brightness profiles are given in Fig. 1 (click here); as can be
seen from the figure, there is a plateau between 2 and 4 Kpc, followed
by the typical disk exponential behaviour.
It is possible to
classify these surface brightness profiles as Freeman II ones. The disk
scalelengths, calculated according to Eq. (1), are given in the first row of
Table 3 (click here).
The scalelengths in B, V, R are almost costant, with a significative increase
in the I band; this behaviour
supports the hypothesis of a significative amount of dust inside the
galactic disk, because of the wavelength dependent absorption. However,
the measurement in the U band invalidates this hypothesis, if one
doesn' t invoke the red-leak effect in the U filter. From Fig. 3 (click here) one
can see that the luminosity profile is almost completely dominated
from the disk fit (1), 
, which implies the impossibility of calculating the bulge
component and the scalelength except for the I band, (see the second row
of Table 3 (click here)).
Figure 3: Radial surface brightness profile in the V
band for NGC 3184 and,
superimposed, the disk fit
The calculated I bulge scalelength is 0.48 Kpc, with a D/B parameter of 4.68; the resulting fit is given in Fig. 4 (click here).
Figure 4: Radial surface brightness profile in the I
band for NGC 3184.
Superimposed, the disk
fit, the bulge fit and the composite trend
disk + bulge
Trying to obtain a better fit of the data, we have tried to fit the disk trend using Eq. (2),
, with 
.
The results are given in Fig. 5 (click here).
Figure 5: As in Fig. 4 (click here),
but the disk fit formula is given by Eq. (2)
As can be seen, the (2) and (3) give a good fit for the disk and bulge components separately, but the composite disk + bulge fit disagrees with data. The bulge scalelength obtained with this fit is very large, 10.17 Kpc in the V band, while the disk scalelength is 2.18 Kpc, with a D/B parameter of 1. The radial disk and bulge scalelengths obtained from (2) and (3) are given respectively in the third and fourth row of Table 3 (click here). All the scalelengths are given in Kpc.
A comparison between Figs. 4 (click here) and 5 (click here) highlights the improved agreement between fits and data obtained using the fits (1) and (3) than that obtained with (2) and (3). The colour indexes are given in Figs. 6 (click here)b.
Figure 6: a) Radial trend of the colour indexes
(B-V), (B-I) with
galactocentric radius
Both in the (B-V) and (B-I) there is a drop for small radii, which proves the existence of a galactic infall, according to Larson & Tinsley (L&T78); a similar behaviour has been detected in NGC 628 (see Paper I). There is also a pronounced depression between 10 and 12 Kpc in (B-I), while (B-V) is smoother, but the noise does not allow to obtain a clear physical intepretation of this fact.
Figure 6: b) Radial trend of the colour index (U-B) with
galactocentric radius
NGC 3423: As can be seen from Fig. 1 (click here), the radial surface brightness profiles show a rapid change in slope at 3 Kpc, while between 0 and 3 Kpc the bulge profile is poorly marked, with almost no fluctuations. The profiles show a trend similar to that of NGC 3184 in the bulge to disk transition zone, so we have attempted to fit the disk using Eq. (2). The results are better than those obtained in the former case, with a I bulge scalelength of 1.15 Kpc, but there is a disagreement between the data and the composite fit disk + bulge for radii greater than 7 Kpc, (see Fig. 7 (click here)).
Figure 7: Radial surface
brightness profile in the I band for NGC 3423.
Superimposed, the disk fit, the bulge
fit and the composite trend disk + bulge.
The disk fit formula is given by Eq. (2), see text
The measured bulge scalelength is consistent with the
galactic profile, but there is no plausible explanation for the
disagreement between data and the fit, except that this deviation
takes place in the outer zone, where the uncertainties in the data are
larger. If Eq. (1) is used for the disk, the disk scalelengths in
the first row of Table 4 (click here) are obtained, while the bulge and its scalelength
can be measured only in the I band.
In the third and fourth rows of the same
table are reported the 
and 
values obtained using Eq. (2)
for the disk.
Figure 8 (click here) shows the luminosity profiles fitted by Eq. (1) and Eq. (3); as
it can be noted, there is a discrepancy between the total disk + bulge
fit and the measured profiles.
This discrepancy in the outer zone of the galactic disk is of the same order
as that obtained
using the Eq. (2) and Eq. (3), while the two fits are
indistinguishable between 0 and 6 Kpc.
Figure 8: As in Fig. 7 (click here),
but the disk fit formula is given by (1)
A useful constraint for choosing the fit may be the D/B parameter; in fact, the latter must be large enough to justify the Tully clasification of this galaxy as Scd. The values obtained for D/B are given in the fifth and sixth rows of Table 4 (click here); there is a dominant contribution from the bulge (which means small D/B) when the composite fit (2) + (3) is used, while the value obtained when the (1) + (3) fit is used, is the one corresponding to a late type galaxy, Sc, so we choose the Eq. (1) as the right fit for the disk.
A comparison between our disk scalelengths in the V and I bands with those obtained by Ryder & Dopita (RD) shows that the disk scalelength we calculated has a bigger value in the I band than in the V one, while the opposite is true in their case; moreover the measured values are different: 2.1 Kpc for the V band and 1.9 Kpc for the I band, according to RD, and 1.16 Kpc for the V band and 1.47 Kpc for the I band according to this work. This difference cannot be ascribed to a different choice of the galactic distance, because both we and RD have used the Tully catalog of nearby galaxies (Tully 1980). Our surface brightness profiles have a behaviour less smooth than those obtained by RD; moreover, their profiles seem to have a double slope in the disk zone, so, the discrepancy between the two disk scalelengths may arise from a different choice of the radial interval in which the exponential fit is applied. The scalelengths calculated from Eq. (2) have less internal dispersion. The colour indexes are given in Figs. 9 (click here)b; it is possible to see a strong reddening in (B-V) and (B-I) toward small radii.
Figure 9: a) Radial trend of the colour indexes (B-V), (B-I) with
galactocentric radius
There is an interesting drop in both (B-I) and (B-V) between 5.5 and 6 Kpc, which may reveal a young stellar population in the outer edge of the galaxy.
Figure 9: b) Radial trend of the colour index (U-B) with
galactocentric radius
The profiles are not really much significative for radii larger than 7-8 Kpc, because of the faintness of the object and the poor S/N.
NGC 3938: For this object the surface brightness radial profiles were measured only in the B, V, I bands (see Fig. 1 (click here)) due to bad weather conditions. In the B band it has not been possible to calculate the bulge scalelength because of the disk domination in the inner zone of the galaxy. The disk and bulge scalelengths are given in Table 5 (click here), together with the values of the D/B parameter.
From the table, one can see that the
larger variations in the disk scalelengths are in the B and V band;
moreover, there is a decreasing trend in 
toward shorter wavelengths
when Eq. (1) is used, while the values are more or less constant when
Eq. (2) is used. The great values obtained for the D/B parameter give this
galaxy a real late type classification, Sd, the Tully ones being Sc.
The small bulge scalelengths in the V and I bands and the impossibility
of measuring it in the B band confirm the idea of a small bulge and then
quite a big D/B. Figure 10 (click here) shows the I band surface brightness
radial profile and, superimposed to it, the composite fit disk plus
bulge coming from the (1) + (3); the colour
indexes (B-V) and (B-I) are shown in Fig. 11 (click here).
Figure 10: The I
band surface brightness radial profile for NGC 3938 and,
superimposed,
the disk fit, the bulge one and the composite trend disk + bulge.
The disk fit formula is given by Eq. (1), see the text
Figure 11: Radial trend of the colour indexes (B-V), (B-I) with
galactocentric radius
As can be seen from Fig. 11 (click here), there is a pronounced reddening toward small radii for both indexes, the same as for NGC 3423. For bigger radii the indexes show a bluish trend, due to the existence of young stellar populations with different ages in the outer disk zone (L&T78).
NGC 6946:
This galaxy has been extensively observed not only in the optical band,
but also in the millimetric band, to study the CO regions by analyzing
the rotational transition 
, at 2.6 mm
(Young & Scoville 1982),
and in the radio band, to study the
HI emision line.
Photometric measures are difficult because of the great number of
field stars, which must be eliminated before evaluating
surface brightness profiles, and because of the large
inclination angle. To obtain luminosity profiles, we have taken
into account both of these factors, by eliminating field stars from the
science frame and by integrating along ellipses in the
observer's plane. The profiles show many bumps, with a sudden change
in the disk slope at 6 Kpc, (see Fig. 1 (click here)), and a drop in luminosity in
the bulge-to-disk transition zone in the U, B, V bands, which is
not present in the R and I ones. There is another change in the
disk slope at 11.5 Kpc, but it is not clear if it is due to an
incomplete measure of galaxy luminosity or to a bad sky subtraction.
The bulge and disk scalelengths, in Kpc, are given in Table 6 (click here), together with
the D/B parameter, while in Fig. 12 (click here) the B surface brightness
radial profile is reported and, superimposed, the fit for the disk and
the bulge.
Figure 12: The B
band surface brightness radial profile for NGC 6946 and,
superimposed,
the disk fit, the bulge fit and the composite trend disk + bulge.
The disk fit formula is given by Eq. (1), see text
The scalelengths are in disagreement with those
obtained from other authors, (see, for example, Young & Scoville
1982), because the different choice of galactic distance, 10
Mpc according Young and Scoville and 5.5 Mpc according Tully (the value
adopted in this work); the measurements of the luminosity profiles are indeed
consistent with these two authors, including the 11.5 Kpc knee.
Disk scalelengths are shorter toward longer
wavelengths, which means that each new stellar population makes the
scalelength increase (see RD); on the contrary, if there were no variation
of 
with wavelength, each new stellar generation would follow the pattern
established by the former ones. The values obtained for the D/B
parameter are in good agreement both with the galaxy classification
by Tully and that from van der Bergh. Phillips et al.
(Phillips et al.
1991), have hypothesized
that this galaxy has an optically thick disk, and have corrected
for autoabsorption using the DDP model and data from Young &
Scoville.
According to these authors, the disk becomes optically thin at
B, so we have attempted to fit the disk using its outer
zone, (7.5-11.5 Kpc; see Fig. 13 (click here)), the profile is completely dominated
by the disk. The disk scalelengths obtained using this radial
interval are given in the last row of Table 6 (click here); as can be seen from
the table, there is not
much difference from those obtained using the whole disk, except
in the R and I bands where the difference is 0.5 Kpc.
Figure 13: The B band surface
brightness radial profile for NGC 6946 and,
superimposed,
the disk fit performed in the radial interval 7.5-11.5 Kpc.
The disk fit formula is given by Eq. (1), see the text
This result, according to DDP, seems to confirm a huge presence of gas and dust inside the galaxy, already indicated by the luminosity drop in surface brightness radial profiles. The disk scalelengths shown in the last row of Table 6 (click here) have a decreasing trend toward longer wavelengths, that seems to confirm the DDP results, but it is in good agreement also with the RD work. The colour indexes are given in Figs. 14 (click here)b.
Figure 14: a)
Radial trend of the colour indexes
(B-I), (B-V) with
galactocentric radius
Figure 14: b) Radial trend of the colour indexes (U-B) with
galactocentric radius
The indexes are superimposed for the whole radial range considered, and have a scaling ratio 1 : 2, while (U-B) seems to have a specular behaviour with respect to the former indexes.
NGC 4321: This is another peculiar galaxy, which is interacting with a companion. However, data analysis has not been modified to take into account the companion, so there is a non negligible influence of its presence in all photometric results, even if the surface brightness luminosity profiles in the B, V, R, I bands (there are no U band measurements due to technical problems during the observing night) are not very disturbed with respect to those of NGC 5194 (M 51), except for the presence of a drop in luminosity in the bulge to disk transition zone. This luminosity drop is more pronounced in the B band than in the I one, (see Fig. 1 (click here)), as well as the luminosity fluctuation in the disk dominated zone of the profile. This is not unexpected, because according to numerical simulation models, (especially by Toomre 1972), the interaction with a companion perturbs the galaxy, changing its matter distribution and triggering a burst of star formation, which is detected mainly in the B band. In the same band, the transitions between arm and interarm zones are more pronounced. In Table 7 (click here) the disk and bulge scalelengths are given, in Kpc, together with the D/B parameter.
Both scalelengths and D/B are consistent with the galaxy classification given by Tully, despite the interaction. There is a smooth trend of increasing disk scalelength toward the infrared, probably due to the presence of dust, which has a very patchy distribution due to the interaction. The colour indexes are given in Figs. 15 (click here)b; they show a very irregular trend, with significative fluctuation also for small radii.
Figure 15: a) Radial trend of the colour index (B-V) with
galactocentric radius
Figure 15: b) Radial trend of the colour indexes (B-I) with
galactocentric radius
It is to be noted that the reddening in both indexes in correspondence of the luminosity drop in the surface brightness profiles is due to, (according to Prieto et al. 1992), the presence of a huge quantity of gas and dust. The presence of a bluish trend of the colour indexes toward small radii, 0-2.5 Kpc seems to confirm the presence of galactic infall, (Larson 1976). The luminosity drop in the bulge to disk transition zone has led Phillips et al. (1991) to consider the disk of this galaxy as optically thick; however, as the galaxy is tidally disturbed and the same authors in a former work (DDP) state that interacting galaxies need a dedicated modellistic for dust distribution, we have not attempted to perform their analysis in this work.
NGC 5194:
The third interacting galaxy is M 51.
For this object we have slightly modified our data
analysis, using circular sectors with an angular width of 
, instead of circular rings, to avoid the luminosity
contamination from the companion; in spite of this, the surface
brightness radial profiles and the colour indexes have turned out to
suffer from the
presence of the interacting companion. The profiles show many bumps,
(see Fig. 1 (click here)), with a sudden change in the disk slope at 6.4 Kpc and a
smoother trend between 2 and 6.4 Kpc; both features are much more
pronounced in the B than in the infrared band. This can be due to the
presence of bursts of star formation induced by the interaction
(Toomre 1972). Nevertheless, it has been possible to fit the disk by
using Eq. (1);
the calculated values for the radial scalelengths, in Kpc, are
given in Table 8 (click here), with those for the bulge in
all five photometric bands, even if the fits are only indicative of
the presence of bulge and disk components and not a real approximation
of the measured profiles, due to the profile bumps (see Fig. 16 (click here)).
Figure 16: The I band surface
brightness radial profile for NGC 5194 and,
superimposed,
the disk fit, the bulge one and the composite trend disk + bulge.
The disk fit formula is given by Eq. (1), see the text
As can be seen from Table 8 (click here), the D/B parameter decreases toward longer wavelengths, with the exception of the U band, which confirms on the one hand the presence of a huge amount of young stars in the disk, and on the other the interaction-induced starburst hypothesis. Disk scalelengths, on the contrary, are more or less constant with various wavelengths. At 10 Kpc, another sudden change in the slope of the disk profile, again due to the companion takes place. A full comparison between our data and those from Boroson (1981), is not possible, because Boroson's profiles are measured only up to 5 - 6 Kpc; however, in this range, there is a good agreement between his profiles and ours. The colour indexes are given in Figs. 17 (click here)b; note in (B-V) and (B-I) a reddening in correspondence with the slope change in the disk zone and a specular bluish in the same zone for the (U-B) index.
Figure 17: a) Radial
trend of the colour indexes (B-I), (B-V) with
galactocentric radius
Figure 17: b) Radial trend of the colour indexes (U-B) with
galactocentric radius
It is possible that the red peak in (B-V) and in (B-I) is due to a dust concentration (Prieto et al. 1992) originated from the tidal interaction; this may be confirmed by a second, less pronounced peak between 1.5 and 3 Kpc. For small radii both (B-V) and (B-I) are bluish; it is not clear if this fact is due to the perturbation (infall) or to an inner spiral arm. There is also a 1 : 2 scaling ratio between the (B-V) and (B-I) indexes.
, 
, D/B for NGC 1232
, 
, D/B for NGC 3184
, 
, D/B for NGC 3423
, 
, D/B for NGC 3938
, 
, D/B for NGC 6946
, 
, D/B for NGC 4321
, 
, D/B for NGC 5194