We aimed to take advantage of the large field of the camera to observe the galaxies at larger radii than hitherto feasible, and thus improve the accuracy of colour gradients. The availability of the series of aperture photometry in PP88 and PN94 for most of the sample was also considered an asset towards a more coherent system of colours. It appears indeed that the colour calibrations are improved here compared to previous work, if this can be judged from the quality of correlations between zero point colours in various surveys (see Sect. 4.1). Two steps in the reduction procedure were thought significant in improving the quality of colour profiles: the first was the adjustment of the FWHM of the PSFs in a given colour set of 5 frames to the best of the five. This allowed us to get significant colours much closer to the galaxy center than otherwise feasible. The second was a careful "mapping'' of the background of each frame, in order to lessen the background fluctuations remaining after the usual flat-fielding procedures. Both these precautions proved successful, and, as a result, the radial range of satisfactory colour measurements was greatly enlarged. Near the galaxy center, it proved feasible to obtain "central colours'', i.e. colours integrated in the circle $r=3\hbox{$^{\prime\prime}$ }$ , in fair agreement with high resolution data (see Sect. 3.2.7.2 and Table 10).

On the other hand, colours could be obtained at much lower surface brightness (or larger radii) than in previous work. Our colour data extend to $23.2 <\mu_V <25.2$ , with a median value near $\mu_V=24.5$ in all colours. According to the comparisons in Sect. 4.1, this is 1.5 to 2 magnitudes deeper than in previous work. "External colours'', refering to the level $\mu_V=24.5$ whenever possible, are published for the first time (see Table 11), and may be useful to give some indications about stellar populations at the outskirts of E-galaxies.

Table 10: Columns 1 to 5: integrated colours within a radius r=3 $^{\prime \prime }$ , i.e. UB₃, BV₃ , VR₃ , VI₃, UV₃; Col. 6: UV_1.5U-V colour at $r=1.5\hbox {$^{\prime \prime }$ }$ calculated from Table 9. Column 7 VR_1.5V-R colour at $r=1.5\hbox {$^{\prime \prime }$ }$ .
NGC UB₃ BV₃ VR₃ VI₃ UV₃ UV_1.5 VR_1.5

2768 0.72 1.04 0.61 1.28 1.76 1.60 0.55

2974 0.58 1.03 - - 1.60 1.72 0.00

3115 0.68 1.03 0.60 1.25 1.72 1.72 0.60

3193 0.60 0.96 0.65 1.32 1.55 1.63 0.58

3377 0.56 0.98 0.54 1.14 1.54 1.41 0.53

3377 0.60 0.98 0.55 1.14 1.57 1.44 0.54

3379 0.72 1.03 0.60 1.23 1.76 1.65 0.60

3605 0.50 0.90 0.57 1.13 1.40 1.43 0.53

3607 0.63 1.04 0.64 1.31 1.67 1.69 0.56

3608 0.61 1.01 0.60 1.24 1.62 1.62 0.56

3610 0.60 0.88 0.55 1.07 1.47 1.44 0.50

3613 0.63 0.98 0.57 1.21 1.61 1.63 0.53

3640 0.58 0.94 0.59 1.21 1.52 1.52 0.56

3872 0.63 0.98 0.62 1.22 1.61 1.64 0.58

4125 0.73 1.06 0.65 1.29 1.79 1.64 0.63

4261 0.73 1.03 0.61 1.29 1.76 1.83 0.60

4278 0.55 0.99 0.60 1.21 1.54 1.53 0.64

4283 0.56 0.96 0.60 1.18 1.53 1.62 0.57

4365 0.71 1.04 0.61 1.28 1.74 1.71 0.64

4374 0.63 1.03 0.64 1.30 1.66 1.60 0.59

4387 0.51 0.98 0.57 1.16 1.49 1.46 0.59

4406 0.59 1.02 0.57 1.26 1.61 1.65 0.58

4472 0.70 1.02 0.61 1.29 1.72 1.78 0.60

4473 0.60 0.95 0.64 1.29 1.55 1.57 0.62

4478 0.45 0.89 0.55 1.17 1.33 1.48 0.55

4486 0.55 0.98 0.63 1.26 1.53 1.84 0.63

4494 0.65 0.93 0.57 1.21 1.57 1.50 0.54

4550 0.39 0.95 0.58 1.22 1.34 1.54 0.62

4551 0.59 0.99 0.58 1.23 1.58 1.51 0.54

4552 0.67 1.07 0.61 1.29 1.74 1.73 0.59

4564 0.63 0.95 0.61 1.19 1.58 1.63 0.57

4621 0.72 1.00 0.63 1.27 1.72 1.74 0.59

4636 0.67 1.00 0.60 1.29 1.66 1.79 0.61

4649 0.71 1.04 0.62 1.29 1.75 1.81 0.62

5322 0.62 0.91 0.62 1.21 1.54 1.53 0.53

5576 0.53 0.85 0.58 1.17 1.38 1.43 0.53

5813 0.64 1.01 0.65 1.33 1.65 1.55 0.61

5831 0.59 0.95 0.57 1.25 1.54 1.50 0.55

5846 0.69 1.01 0.62 1.28 1.71 1.77 0.61

5866 - - - - - 1.60 0.61

5982 0.59 0.94 0.54 1.18 1.53 1.53 0.58

**Table 10:** Columns 1 to 5: integrated colours within a radius r=3 $^{\prime \prime }$ , i.e. UB₃, BV₃ , VR₃ , VI₃, UV₃; Col. 6: UV_1.5U-V colour at $r=1.5\hbox {$^{\prime \prime }$ }$ calculated from Table 9. Column 7 VR_1.5V-R colour at $r=1.5\hbox {$^{\prime \prime }$ }$ .
NGC	UB₃	BV₃	VR₃	VI₃	UV₃	UV_1.5	VR_1.5
2768	0.72	1.04	0.61	1.28	1.76	1.60	0.55
2974	0.58	1.03	-	-	1.60	1.72	0.00
3115	0.68	1.03	0.60	1.25	1.72	1.72	0.60
3193	0.60	0.96	0.65	1.32	1.55	1.63	0.58
3377	0.56	0.98	0.54	1.14	1.54	1.41	0.53
3377	0.60	0.98	0.55	1.14	1.57	1.44	0.54
3379	0.72	1.03	0.60	1.23	1.76	1.65	0.60
3605	0.50	0.90	0.57	1.13	1.40	1.43	0.53
3607	0.63	1.04	0.64	1.31	1.67	1.69	0.56
3608	0.61	1.01	0.60	1.24	1.62	1.62	0.56
3610	0.60	0.88	0.55	1.07	1.47	1.44	0.50
3613	0.63	0.98	0.57	1.21	1.61	1.63	0.53
3640	0.58	0.94	0.59	1.21	1.52	1.52	0.56
3872	0.63	0.98	0.62	1.22	1.61	1.64	0.58
4125	0.73	1.06	0.65	1.29	1.79	1.64	0.63
4261	0.73	1.03	0.61	1.29	1.76	1.83	0.60
4278	0.55	0.99	0.60	1.21	1.54	1.53	0.64
4283	0.56	0.96	0.60	1.18	1.53	1.62	0.57
4365	0.71	1.04	0.61	1.28	1.74	1.71	0.64
4374	0.63	1.03	0.64	1.30	1.66	1.60	0.59
4387	0.51	0.98	0.57	1.16	1.49	1.46	0.59
4406	0.59	1.02	0.57	1.26	1.61	1.65	0.58
4472	0.70	1.02	0.61	1.29	1.72	1.78	0.60
4473	0.60	0.95	0.64	1.29	1.55	1.57	0.62
4478	0.45	0.89	0.55	1.17	1.33	1.48	0.55
4486	0.55	0.98	0.63	1.26	1.53	1.84	0.63
4494	0.65	0.93	0.57	1.21	1.57	1.50	0.54
4550	0.39	0.95	0.58	1.22	1.34	1.54	0.62
4551	0.59	0.99	0.58	1.23	1.58	1.51	0.54
4552	0.67	1.07	0.61	1.29	1.74	1.73	0.59
4564	0.63	0.95	0.61	1.19	1.58	1.63	0.57
4621	0.72	1.00	0.63	1.27	1.72	1.74	0.59
4636	0.67	1.00	0.60	1.29	1.66	1.79	0.61
4649	0.71	1.04	0.62	1.29	1.75	1.81	0.62
5322	0.62	0.91	0.62	1.21	1.54	1.53	0.53
5576	0.53	0.85	0.58	1.17	1.38	1.43	0.53
5813	0.64	1.01	0.65	1.33	1.65	1.55	0.61
5831	0.59	0.95	0.57	1.25	1.54	1.50	0.55
5846	0.69	1.01	0.62	1.28	1.71	1.77	0.61
5866	-	-	-	-	-	1.60	0.61
5982	0.59	0.94	0.54	1.18	1.53	1.53	0.58

Table 11: ``External colours'' measured in the outermost range of the data. Successive columns: $\mu _V$ V surface brightness of measurement, a range of 0.5 mag centered at the quoted $\mu _V$ beeing used. Columns 2-5: U-B, B-V, V-R, V-I respectively with estimated errors. These are the same in V-R as in B-V.
NGC $\mu _V$ U-B B-V V-R V-I

2768 24.25 $0.455\pm.06$ $0.851\pm.02$ 0.530 $1.110\pm.04$

2974 id $0.335\pm.06$ $0.835\pm.03$ 0.000 $0.000\pm.00$

3115 id $0.365\pm.05$ $0.855\pm.02$ 0.580 $1.180\pm.03$

3193 23.5 $0.370\pm.04$ $0.880\pm.02$ 0.000 $0.000\pm.00$

3377 24.5 $0.210\pm.06$ $0.815\pm.02$ 0.505 $1.000\pm.03$

3377 24.25 $0.175\pm.05$ $0.795\pm.02$ 0.480 $0.950\pm.03$

3379 24.0 $0.480\pm.04$ $0.890\pm.02$ 0.575 $1.160\pm.03$

3605 23.5 $0.340\pm.03$ $0.795\pm.02$ 0.510 $1.060\pm.02$

3607 24.25 $0.370\pm.07$ $0.845\pm.03$ 0.530 $1.130\pm.04$

3608 23.75 $0.300\pm.05$ $0.900\pm.02$ 0.525 $1.165\pm.03$

3610 24.25 $0.365\pm.05$ $0.730\pm.02$ 0.510 $0.985\pm.04$

3613 24.5 $0.395\pm.07$ $0.830\pm.02$ 0.500 $1.130\pm.04$

3640 23.5 $0.360\pm.04$ $0.875\pm.02$ 0.540 $1.135\pm.03$

3872 24.5 $0.420\pm.07$ $0.890\pm.03$ 0.555 $1.110\pm.05$

4125 23.5 $0.405\pm.03$ $0.830\pm.01$ 0.550 $1.120\pm.02$

4278 23.5 $0.350\pm.04$ $0.800\pm.02$ 0.530 $1.055\pm.03$

4283 23.75 $0.300\pm.05$ $0.805\pm.02$ 0.510 $1.080\pm.03$

4365 24. $0.470\pm.04$ $0.900\pm.02$ 0.585 $1.180\pm.03$

4374 23.25 $0.420\pm.03$ $0.895\pm.01$ 0.545 $1.180\pm.03$

4387 24.5 $0.350\pm.08$ $0.860\pm.03$ 0.550 $1.140\pm.05$

4406 23.25 $0.330\pm.04$ $0.870\pm.02$ 0.565 $1.160\pm.03$

4472 23. $0.490\pm.03$ $0.930\pm.01$ 0.590 $1.255\pm.02$

4473 24.5 $0.315\pm.08$ $0.840\pm.03$ 0.575 $1.130\pm.05$

4478 24. $0.200\pm.04$ $0.750\pm.02$ 0.490 $1.090\pm.03$

4486 23.5 $0.415\pm.03$ $0.885\pm.02$ 0.590 $1.200\pm.03$

4494 24.5 $0.390\pm.08$ $0.830\pm.03$ 0.500 $1.110\pm.05$

4550 24.0 $0.120\pm.06$ $0.700\pm.02$ 0.470 $1.010\pm.03$

4551 24.5 $0.325\pm.07$ $0.850\pm.03$ 0.520 $1.120\pm.04$

4552 24. $0.390\pm.06$ $0.880\pm.02$ 0.515 $1.120\pm.03$

4564 24.5 $0.200\pm.07$ $0.820\pm.02$ 0.530 $1.020\pm.03$

4621 24. $0.390\pm.06$ $0.370\pm.02$ 0.565 $1.175\pm.03$

4636 23.5 $0.370\pm.03$ $0.780\pm.02$ 0.510 $1.140\pm.03$

4649 23.5 $0.525\pm.04$ $0.950\pm.02$ 0.540 $1.200\pm.03$

5322 24.25 $0.315\pm.07$ $0.785\pm.03$ 0.520 $1.110\pm.04$

5576 23.75 $0.230\pm.05$ $0.770\pm.02$ 0.525 $1.085\pm.03$

5813 24. $0.410\pm.06$ $0.910\pm.02$ 0.575 $1.175\pm.03$

5831 23. $0.365\pm.06$ $0.940\pm.02$ 0.580 $1.225\pm.03$

5866 24.5 $0.270\pm.08$ $0.840\pm.02$ 0.495 $1.115\pm.04$

5846 24. $0.420\pm.06$ $0.940\pm.02$ 0.580 $1.225\pm.03$

5866 24.5 $0.270\pm.08$ $0.775\pm.03$ 0.505 $1.075\pm.05$

5982 24.25 $0.365\pm.07$ $0.820\pm.03$ 0.500 $1.100\pm.04$

**Table 11:** ``External colours'' measured in the outermost range of the data. Successive columns: $\mu _V$ V surface brightness of measurement, a range of 0.5 mag centered at the quoted $\mu _V$ beeing used. Columns 2-5: U-B, B-V, V-R, V-I respectively with estimated errors. These are the same in V-R as in B-V.
NGC	$\mu _V$	U-B	B-V	V-R	V-I
2768	24.25	$0.455\pm.06$	$0.851\pm.02$	0.530	$1.110\pm.04$
2974	id	$0.335\pm.06$	$0.835\pm.03$	0.000	$0.000\pm.00$
3115	id	$0.365\pm.05$	$0.855\pm.02$	0.580	$1.180\pm.03$
3193	23.5	$0.370\pm.04$	$0.880\pm.02$	0.000	$0.000\pm.00$
3377	24.5	$0.210\pm.06$	$0.815\pm.02$	0.505	$1.000\pm.03$
3377	24.25	$0.175\pm.05$	$0.795\pm.02$	0.480	$0.950\pm.03$
3379	24.0	$0.480\pm.04$	$0.890\pm.02$	0.575	$1.160\pm.03$
3605	23.5	$0.340\pm.03$	$0.795\pm.02$	0.510	$1.060\pm.02$
3607	24.25	$0.370\pm.07$	$0.845\pm.03$	0.530	$1.130\pm.04$
3608	23.75	$0.300\pm.05$	$0.900\pm.02$	0.525	$1.165\pm.03$
3610	24.25	$0.365\pm.05$	$0.730\pm.02$	0.510	$0.985\pm.04$
3613	24.5	$0.395\pm.07$	$0.830\pm.02$	0.500	$1.130\pm.04$
3640	23.5	$0.360\pm.04$	$0.875\pm.02$	0.540	$1.135\pm.03$
3872	24.5	$0.420\pm.07$	$0.890\pm.03$	0.555	$1.110\pm.05$
4125	23.5	$0.405\pm.03$	$0.830\pm.01$	0.550	$1.120\pm.02$
4278	23.5	$0.350\pm.04$	$0.800\pm.02$	0.530	$1.055\pm.03$
4283	23.75	$0.300\pm.05$	$0.805\pm.02$	0.510	$1.080\pm.03$
4365	24.	$0.470\pm.04$	$0.900\pm.02$	0.585	$1.180\pm.03$
4374	23.25	$0.420\pm.03$	$0.895\pm.01$	0.545	$1.180\pm.03$
4387	24.5	$0.350\pm.08$	$0.860\pm.03$	0.550	$1.140\pm.05$
4406	23.25	$0.330\pm.04$	$0.870\pm.02$	0.565	$1.160\pm.03$
4472	23.	$0.490\pm.03$	$0.930\pm.01$	0.590	$1.255\pm.02$
4473	24.5	$0.315\pm.08$	$0.840\pm.03$	0.575	$1.130\pm.05$
4478	24.	$0.200\pm.04$	$0.750\pm.02$	0.490	$1.090\pm.03$
4486	23.5	$0.415\pm.03$	$0.885\pm.02$	0.590	$1.200\pm.03$
4494	24.5	$0.390\pm.08$	$0.830\pm.03$	0.500	$1.110\pm.05$
4550	24.0	$0.120\pm.06$	$0.700\pm.02$	0.470	$1.010\pm.03$
4551	24.5	$0.325\pm.07$	$0.850\pm.03$	0.520	$1.120\pm.04$
4552	24.	$0.390\pm.06$	$0.880\pm.02$	0.515	$1.120\pm.03$
4564	24.5	$0.200\pm.07$	$0.820\pm.02$	0.530	$1.020\pm.03$
4621	24.	$0.390\pm.06$	$0.370\pm.02$	0.565	$1.175\pm.03$
4636	23.5	$0.370\pm.03$	$0.780\pm.02$	0.510	$1.140\pm.03$
4649	23.5	$0.525\pm.04$	$0.950\pm.02$	0.540	$1.200\pm.03$
5322	24.25	$0.315\pm.07$	$0.785\pm.03$	0.520	$1.110\pm.04$
5576	23.75	$0.230\pm.05$	$0.770\pm.02$	0.525	$1.085\pm.03$
5813	24.	$0.410\pm.06$	$0.910\pm.02$	0.575	$1.175\pm.03$
5831	23.	$0.365\pm.06$	$0.940\pm.02$	0.580	$1.225\pm.03$
5866	24.5	$0.270\pm.08$	$0.840\pm.02$	0.495	$1.115\pm.04$
5846	24.	$0.420\pm.06$	$0.940\pm.02$	0.580	$1.225\pm.03$
5866	24.5	$0.270\pm.08$	$0.775\pm.03$	0.505	$1.075\pm.05$
5982	24.25	$0.365\pm.07$	$0.820\pm.03$	0.500	$1.100\pm.04$

On the other hand, the "red halo'' effect of the camera was found to give enormous errors in V-I colours and gradients. These were corrected by a rigorous technique, and results in agreement with "classical'' data were obtained. Considering the V-I gradients, one is not happy however to introduce in their evaluation, corrections larger than the quantity to be measured! Besides this specific problem of the red halo of thinned CCD, the far wings of the PSFs have been proven in a recent paper (RM01) to have non negligible effects in the gradients of other colours, and also to vary with the age of mirror coatings. It is not impossible that the U-B or U-V gradients given here are overestimated by 15-20%, although they are in excellent statistical agreement with the well-known work of Peletier et al. (1990).

Various colour gradients against $\log r$ for a given object are well correlated, generally better than in previous work (see statistics in Sect. 4.1), which is interpreted as due to smaller measuring errors, notably in U-B. These improvements in accuracy did not bring out any obvious correlation between gradients and other galaxy properties. A few galaxies have exceptionally steep colour gradients (nearly at $2\sigma$ ) without sharing other properties.

Colour-colour relations can be built from the present data for several locations in galaxies, such as near center, various fractions of the effective radius $r_{\rm e}$ , or the "outermost'' measured range around $\mu _V=24$ . All these diagrams overlap to form a single stripe with moderate scatter (except for one rather obvious calibration error?). These might prove useful to test theories of old stellar populations and of their host galaxies. Colour-colour diagrams based upon integrated colours have already been used for this purpose (Worthey 1994).

The U-B or U-V colours correlate very well with the Mg₂ index, both near the galaxy center and at the effective radius $r_{\rm e}$ . This seems to rule out any large influence of diffuse dust in the colours and colour gradients in E-galaxies. This was considered likely by Witt et al. (1992) and discussed by Wise & Silva (1996) with inconclusive results. Previous arguments against such an influence were presented in RM00: they were based upon the relative average values of the gradients in various colours, and are reinforced in the present work, since the mean gradients are nearly unchanged, and their errors lessened.

It is well known that, for single-burst stellar populations, colours and line indices depend both on the metallicity and on the age of the system (Worthey 1994; Borges et al. 1995). However, E-galaxies are constituted by a population mix, having age and metallicity distributions which reflect their star formation histories. Therefore a colour-metallicity calibration requires the use of models able to provide those distributions and, consequently, the integrated colours along the galaxy lifetime. Such a calibration will be presented in Paper II.

5 Conclusions