3 The data

We have obtained K' images for the 18 active galaxies together with Jimages for 15 of them. J and K' images were taken for 11 non-active spirals. The infrared data have been acquired mainly with three telescopes: the 2 m Télescope Bernard Lyot (Pic du Midi Observatory, France), the 2.5 m DuPont telescope (Las Campanas, Chile), and the 3.5 m telescope at Calar Alto (Spain). We have also used some short observations taken with the ESO 2.2 m telescope in La Silla for the relative calibration of one of the J images obtained during a non-photometric night. The journal of observations is presented in Table 2. We always used NICMOS-3 256 $\times$ 256 detectors; the pixel scales are given in Table 2.

   

Table 2: Journal of observations

Name	Teles.	Obs.	Filter	Exp.	Seeing	$\mu_{2\sigma}$
		date		time	FWHM	(mag/
				(s)	(arcsec)	arcsec2)
UGC 1395	CAHA	09/96	K'	5400	0.95	20.7
IC 184	CAHA	09/96	K'	1800	0.95	19.8
IC 1816	LC	09/96	K'	3000	-	20.1
			J	800	-	21.0
UGC 3223	LC		K'	2700	1.2	19.9
NGC 26391	TBL	03/97	J	1000	-	20.5
			K'	3000	-	19.7
IC 2510*	LC	05/96	K'	3600	0.87	19.6
		04/97	J	1000	0.71	20.8
NGC 3281	LC	05/96	K'	3500	0.84	19.8
		04/97	J	1000	0.89	21.1
NGC 3660	LC	04/97	J	1200	-	21.3
		04/97	K'	2000	-	20.2
NGC 42533	TBL	03/97	J	1000	-	21.3
			K'	2800	-	19.3
NGC 4507*	LC	05/96	J	1400	0.80	21.0
			K'	3600	0.97	20.2
NGC 47853	LC	04/97	J	1000	0.71	20.8
			K'	3000	0.71	20.2
NGC 5347*,2	TBL	03/97	J	1000	1.4	21.3
			K'	3060	1.9	19.4
NGC 5728	LC	04/97	J	1000	0.88	20.9
			K'	2800	0.85	20.2
ESO 139-12	LC	05/96	J	1000	1.0	21.2
			K'	3400	0.97	19.5
NGC 68143	LC	05/96	J	1000	1.2	20.9
			K'	3200	0.87	19.3
NGC 68603	LC	04/97	J	1000	0.87	20.9
			K'	2600	0.89	20.3
	ESO	05/96	K'	720	0.89	17.5
NGC 6890	LC	04/97	J	1000	0.76	20.7
			K'	2600	0.78	20.1
NGC 6951	NOT	09/96	J	1760	1.3	18.8
			K'	3520	1.3	19.5
NGC 1511	LC	09/96	J	800	-	21.1
			K'	1800	-	19.9
IC 454	TBL	03/97	J	1000	1.2	20.3
			K'	2800	1.25	19.2
NGC 2712	TBL	03/97	J	1000	1.1	20.7
			K'	3200	1.2	20.1
NGC 2811*	LC	05/96	K'	3400	1.0	19.8
		04/97	J	1000	1.19	20.7
NGC 35711	LC		J	1000	0.84	21.1
			K'	3000	0.67	20.1
NGC 3835*	TBL	03/97	J	1000	-	21.3
			K'	3000	-	19.7
NGC 4162	TBL	03/97	J	800	1.6	21.3
			K'	3000	1.1	20.0
NGC 4290	TBL	03/97	J	1200	1.1	21.0
			K'	3000	1.1	19.5
NGC 4779	TBL	03/97	J	4$\times$1000	1.1	20.9
			K'	4$\times$1000	1.1	19.3
NGC 6012	TBL	03/97	J	1000	1.2	21.0
			K'	3600	1.35	19.7
NGC 6155	TBL	03/97	J	600	1.15	21.0
			K'	3200	1.65	19.5

CAHA: 3.5 m telescope at Calar Alto ( $\rm0.32''\!/pixel$); LC: 2.5 m telescope at Las Campanas ( $\rm0.348''\!/pixel$); TBL: 2 m Bernard Lyot telescope at Pic du Midi ( $\rm0.5''\!/pixel$); ESO: 2.2 m telescope at La Silla ( $\rm0.278''\!/pixel$); NOT: Nordic Optical Telescope at La Palma. ( $\rm0.51''\!/pixel$).
^* Galaxies with non-flat remaining background.
^1,2,3 Galaxies with NICMOS HST archive images, (1) published by Böker et al. (1999); (2) published by Regan & Mulchaey (1999); (3) unpublished.

As seen in Table 1, there are some galaxies with extended angular sizes close to 2 arcmin or even larger, which do not fit in a single frame. For two of them, namely NGC 4779 and NGC 6951, a mosaic of images was performed in order to have images for the entire galaxy disk.

The data reduction and calibration was performed, following the standard procedures with the IRAF software and SQIID package. For the objects, exposures of typically about 2 minutes (dithered in short exposures of about 20 s in J and 50 s in K' offset by about 2-3 arcsec) were taken on the source and on the sky, until the total integration time was achieved (see Table 2). A median dark frame was calculated for every night with the same exposure as that of the objects by using the corresponding dark frames obtained before and after the observations. Flat-field frames were obtained for each filter by subtracting the median dark frame from the sky frames, normalizing the resulting frames and calculating the median. We then subtracted the dark frame from the object frame, divided by the normalized flat-field and subtracted the corresponding sky frame. Sometimes, another constant had to be subtracted in order to attain the zero background level. In order to align the resulting frames to better than a pixel fraction we used the non-saturated stars in these frames or, when there were no stars, we used the centre of the galaxy. For the flux calibration a number of standard stars were observed and mean extinction coefficients were applied (k_J and k_K' in magnitudes airmass^-1 are respectively 0.12 and 0.08 for Las Campanas and La Silla, 0.12 and 0.10 for the TBL, 0.057 and 0.042 for the NOT and k_K'=0.1 for Calar Alto). The error level on the standard stars amounts to 10%. In Col. 7 of Table 2 we give the isophotal magnitude/arcsec² corresponding to 2$\sigma$ of the background.

We have retrieved HST calibrated infrared images obtained with NICMOS (filter F160W) for the eight galaxies for which such data was available (see Col. 1 in Table 2).