2 Object selection and observations

Our galaxy sample is comprised of 9 edge-on spirals for which we already possess deep images in the H${\alpha }$ emission-line ( $\lambda=6563$ Å). These galaxies were originally selected on the basis of their appearing nearby (apparent diameter 5-10') and having inclinations within a few degrees of edge-on (R96). We added to this NGC 891 which has been extensively studied by one of us (Rand 1998; Rand et al. 1990) and, as already indicated, represents the archetypal or prototypal "chimney galaxy''. It is important to emphasize that our main sample has not been chosen on the basis of either high emission-line brightness or enhanced far-infrared (FIR) luminosity. It is expected then to preclude any bias towards active or starburst galaxies, making our results far-reaching in the sense that they will apply to typical spiral disks. We use both the FIR luminosity and the ratio of FIR-to-blue luminosities to check the level of recent star-formation in our sample. Both quantities are known to assume high values in systems displaying starburst activity (Heckman et al. 1990; Soifer et al. 1997; Alton et al. 1998b). High Resolution IRAS maps (HiRes) at 60 and $100~\mu$m were obtained on request from the Infrared Processing and Analysis Center at Caltech (IPAC). These data, in conjunction with optical photometry from the RC3 (de Vaucouleurs et al. 1991), allowed us to infer mean values of 4.6 10⁹ $L_{\odot}$ and 1.1 for the FIR luminosity and $L_{\rm FIR}$-to-L_B ratio respectively (Table 1). These quantities correspond much better to the properties of nearby, optically-selected IRAS galaxies ($5\ 10^{9}$ $L_{\odot}$ and 0.6 respectively) rather than to starburst disks ( $5\ 10^{10}$ and $\sim 10$ respectively), and confirms that our sample consists chiefly of quiescent rather than starburst galaxy types (Soifer et al. 1987; Heckman et al. 1990). We can be less sure that NGC 891 complies with the original selection criteria. It is indeed one of the nearest examples of a large edge-on disk but an extensive history of observation indicates that it possesses a fairly active halo which is probably fed by a disk harbouring fairly vigorous star-formation. The FIR properties of NGC 891 indicate somewhat enhanced levels of recent star-formation over "normal'' disks but Table 1 shows that this galaxy is much better classified as a quiescent rather than a starburst system.

 

  Table 1: Basic properties of the galaxy sample. The object position, spiral type and distance are taken from Rand (1996). The blue magnitude, $B_{\rm corr}$, is extracted from the RC3 and is corrected for Milky Way extinction only (i.e. no account of internal extinction has been taken). $L_{\rm FIR}$ is the far-infrared luminosity obtained from integrating 60 and $100~\mu$m flux densities in the HiRes IRAS data. The FIR-to-blue luminosity ratio follows the conventional definition of Rice et al. (1988)

Galaxy	Other Designation	Type	R.A.	Dec.	D	$B_{\rm corr}$	$L_{\rm FIR}$	$\frac{L_{\rm FIR}}{L_{\rm B}}$
			(J2000)	(J2000)	(Mpc)	(mag)	(1010$L_{\odot}$)
NGC 891	UGC 01831	Sb	02 22 33.0	+42 20 56	9.5	10.5	1.4	3.2
UGC 4278	IC 2233	Sd	08 13 58.8	+45 44 36	10.6	12.9	0.014	0.23
NGC 4013	UGC 06963	Sbc	11 58 31.1	+43 56 50	17.0	12.2	0.47	1.6
NGC 4217	UGC 07282	Sb	12 15 50.7	+47 05 37	17.0	12.0	0.80	2.4
NGC 4302	UGC 07418	Sc	12 21 42.3	+14 35 52	16.8	12.4	0.22	0.93
NGC 4762	UGC 08016	SO	12 52 56.3	+11 13 48	16.8	11.1	<0.025	<0.033
NGC 5023	UGC 08286	Scd	13 12 12.0	+44 02 20	6	12.9	0.0049	0.25
NGC 5746	UGC 09499	Sb	14 44 56.4	+01 57 16	29.4	11.2	0.66	0.29
NGC 5907	UGC 09801	Sc	15 15 54.0	+56 19 45	14.9	11.1	0.76	1.25
UGC 10288	-	Sc	16 14 25.0	-00 12 25	31.5	13.8	0.26	1.13

 

 

Table 2: B-band observations for the galaxy sample

Galaxy	Telescope	Observing date	Exposure	Seeing FWHM
			(min)	(arcsec)
NGC 891	Skinakas 1.3 m	October 1996	80	1.4
UGC 4278	INT 2.5 m	December 1995	10	2.5
NGC 4013	Skinakas 1.3 m	June 1995	140	1.2
NGC 4217	JKT 1.0 m	April 1999	90	1.2
NGC 4302	INT 2.5 m	April 1991	17	1.3
NGC 4762	JKT 1.0 m	March 1994	90	1.6
NGC 5023	JKT 1.0 m	April 1993	60	2.0
NGC 5746	INT 2.5 m	June 1990	8.3	1.8
NGC 5907	Skinakas 1.3 m	May 1997	30	1.8
UGC 10288	JKT 1.0 m	April 1999	90	1.2

In order to carry out a rigorous search for extraplanar extinction features we required deep, optical images for our sample. For a few of the galaxies in Table 1 (NGC 891, NGC 4013, NGC 5907), we were already in possession of deep B, V and R images taken with the 1.3-m telescope at Skinakas Observatory, Greece (Xilouris et al. 1999). For the remainder, we carried out B-band imaging at the Kitt Peak 0.6-m telescope during April 1998 but bad weather prevented us from obtained images with sufficient signal. To complete coverage of the sample we accessed data from the Isaac Newton Group (ING) archive at Cambridge. Table 2 summarizes the sources and important properties of the B-band data. Since our sample was not observed under constant seeing conditions using an identical instrumental set-up we will examine, at a later stage, the effect that varied observing conditions might have on our sensitivity to extraplanar dust.

The archive data were reduced in the standard manner using the STARLINK software package CCDPACK to perform debiasing, flatfielding and sky subtraction etc. We also removed the effects of cosmic rays by aligning and median stacking multiple exposures of the same object. It was not necessary to photometrically calibrate any of the images. Neither the recognition of dust chimneys nor the calculation of optical depth for such features relies on the absolute brightness of the galaxy but solely on the relative difference in brightness between attenuated regions and adjacent, unextinguished regions.