2 The source list and VLBA observations

Among the twenty-eight sources selected for VLBA observations, twenty-four are unresolved or slightly resolved (angular size $\leq$0.2$\arcsec$) in the VLA 8.4 GHz observations. The remaining four sources are dominated in flux density by an unresolved or slightly resolved component, with the additional presence of some weaker emission (bringing the overall size up to $\sim$1$\arcsec$, Table 1) whose connection with the compact feature is unclear. In addition about half of the sample sources present a low frequency turnover in the total spectrum which is usually considered an indication of compactness in some sub-component. All these sources, therefore, deserved further investigation.

The selected sample is presented in Table 1. The content of the table is the following.

Column 1: Source name; an "$\ast$'' refers to a note (Sect. 4);

Column 2: Optical Identification (Id) from Paper I (G = galaxy, Q = quasar, E = no known optical counterpart);

Column 3: R magnitude;

Column 4: redshift; "K'' and "R'' indicate that the redshift is estimated by photometric measurements in the respective optical band (see Paper I for details);

Column 5: VLA Largest Angular Size (LAS) in mas, from Paper I;

Column 6: log $P_{\rm0.4 GHz}$ (P in W/Hz h^-2); for E sources lower limits to the radio power have been computed assuming z = 0.5 (see also Paper I for a wider discussion);

Column 7: flux density (mJy), interpolated at 1.67 GHz from the 1.4 GHz NVSS (Condon et al. 1998) and the 5 GHz VLA data from Paper I;

Column 8: VLBA 1.67 GHz flux density (mJy), measured on the images presented here (see Sect. 3);

Column 9: VLBA Largest Angular Size (mas);

Column 10: Largest Linear Size (LLS, kpc h^-1) from VLBA data; for E sources z = 1.05 was assumed (see Paper I for details) and LLS, given with a leading "$\sim$'', must be considered a lower limit;

Column 11: overall morphology (see Sect. 5).

The VLBA observations were carried out in February 2000 at the frequency of 1.67 GHz, with a recording band-width of 32 MHz at 64 Mbps. Each source was observed for a total of about 1 hour, spread into 10 to 15 short scans in order to improve the uvcoverage. A typical example of uv coverage is given in Fig. 1. As no single VLA antenna was available, the shortest baseline is Pie Town-Los Alamos ($\sim$1.3 M$\lambda$). This may have caused some flux density losses in the more extended sources/components.

 

 

Table 2: Observational parameters of source components.

Source		S	$\theta_1$	$\theta_2$	(PA)	Source		S	$\theta_1$	$\theta_2$	(PA)
		(mJy)	mas	mas	deg			(mJy)	mas	mas	deg
(1)	(2)	(3)	-	(4)	-	(1)	(2)	(3)	-	(4)	-
0039+373	N1	229	4.7	4.3	101	1225+442	E1	138	26.6	15.1	88
	N2	143	11.0	6.8	17		E$^\ast$	28
	N$^\ast$	17					W1	94	46.6	23.3	71
	S1	303	13.2	6.9	24		W2	35	32.4	10.6	134
	S$^\ast$	88				1242+410	S	818	20.8	5.1	26
0147+400	E	208	2.7	2.4	41		N	208	7.7	4.8	53
	W1	134	11.6	7.6	40		$T^\ast$	101
	W2	109	13.7	10.6	26	1314+453	W1	175	26.7	15.6	71
	$T^\ast$	113					W2	77	31.6	26.9	73
0703+468	W1	794	7.1	4.01	51		Ce	60	19.0	15.0	143
	W$^\ast$	22					E1	101	28.0	14.6	63
	E1	497	5.7	4.0	80		E2	76	41.4	10.2	65
	E$^\ast$	44					NW	18	25.0	8.9	82
0800+472	S1	177	25.5	9.4	49	1340+439	N1	208	9.0	5.2	9
	S2	95	32.4	6.9	70		N2	104	13.9	5.8	44
	N1	120	36.5	19.2	34		Ce1	38	10.3	1.5	26
	N2	68	25.6	22.9	60		Ce2	13	7.2	2.2	4
	N3	24	18.8	11.3	178		S	27	12.0	6.6	83
0809+404	N	250	16.1	12.2	161	1343+386	S1	541	6.2	5.9	131
	S	221	17.5	12.2	161		S$^\ast$	68
	$T^\ast$	51					Ce	20	12.8	4.5	170
0822+394	E1	424	5.3	3.5	37		N	66	3.5	0.6	171
	E2	269	16.4	3.3	57		N$^\ast$	20
	W1	153	6.7	4.5	66	1432+428	E1	570	4.9	3.0	106
	W2	97	6.4	3.7	81		E$^\ast$	25
0840+424	N1	823	8.2	4.8	10		W	104	11.4	7.5	133
	N2	145	12.5	4.5	169	1441+409	W1	312	6.7	4.2	50
	S1	71	7.1	5.7	50		W2	132	17.1	8.0	43
	S2	137	16.8	11.6	139		Ce1	140	7.6	3.0	63
1007+422	S1	238	32.9	15.8	84		Ce2	42	<3.5	<3.5	-180
	S2	52	22.4	8.7	40		E1	70	8.0	3.1	176
	N	32	40.0	23.5	153		E2	68	17.2	7.3	30
1008+423	W1	139	12.3	4.4	128	1449+421	N1	121	12.04	7.0	96
	W2	250	12.0	5.3	35		N2	149	5.2	2.7	26
	W$^\ast$	14					N3	87	8.09	4.1	4
	Ce	14	12.2	5.2	54		Ce1	15	8.5	2.6	24
	E	104	27.8	19.4	46		Ce2	58	5.8	3.4	31
1016+443	N1	109	8.1	5.5	156		Ce3	44	8.9	2.1	27
	N2	67	15.1	3.5	175		S	120	11.0	8.2	60
	N3	6	15.9	1.6	166	2304+377	W1	244	4.8	0.2	89
	S1	27	15.1	7.7	155		W2	671	21.0	6.6	117
	S2	58	18.9	7.2	5		W3	91	7.1	2.3	104
1044+454	E1	124	8.3	3.8	106		E1	219	4.0	3.3	114
	E2	138	19.9	17.1	110		E2	41	9.0	5.6	67
	E$^\ast$	18					E3	124	28.0	17.8	110
	W1	24	7.3	4.4	6	2330+402	W1	332	2.4	1.8	33
	W2	7	10.6	1.7	177		W$^\ast$	37
1049+384	W1	244	7.0	4.1	8		E1	62	7.0	4.8	152
	W2	206	8.7	2.5	119		E2	204	17.5	12.9	164
	E1	21	<5.5	<5.5	-179		E$^\ast$	70
	E2	26	5.4	3.6	133	2348+450	S1	203	5.1	4.1	159
1133+432	N	799	3.5	2.9	48		S$^\ast$	168
	S	460	4.1	3.6	167		N1	75	19.9	14.4	93
1136+383	N1	134	5.2	3.0	170		N$^\ast$	74
	N2	93	10.1	4.1	10	2358+406	S1	681	5.6	3.9	143
	Ce	23	8.2	3.2	170		S$^\ast$	158
	S	125	5.8	5.1	3		Ce	71	9.3	1.0	156
1136+420	T	292	52.2	31.1	103		N1	95	8.8	4.3	3
1159+395	N1	198	5.2	3.3	179		N2	213	15.1	6.1	123
	N2	130	21.3	9.4	167
	S1	165	6.7	4.8	141

Figure 2: (left): comparison between the VLBA integrated flux densities and total flux densities interpolated from the literature. The line has a slope of one. (right): ratio between the VLBA and the total flux density is shown as a function of the overall source angular size.