Observations were carried out in two separate VLBI sessions (September 1993 and May 1994) using the 6 antennas listed in Table 1 (click here). The data (taken in right circular polarization, RCP) were recorded in MK3 mode A, allowing a total bandwidth of 56 MHz. The fourteen 4-MHz bands were split between X band (8.3 GHz) with 8 channels (32 MHz) and the S band with 6 channels (24 MHz). Due to hardware limitations, Noto could record only 4 channels at X band in session 1, and again 4 at X band and 2 at S band in session 2.

**Table 2:** Observed sources: Col. 1 shows the IAU name, Col. (2) the optical ID, Col. 3 the redshift from literature data, Col. 4 the radio power at 5 GHz, Col. 5 displays the largest angular size, Col. 6 contains the VLBI classification from our data (P = pointlike; S = single component, resolved either at S or at X band, or both; D = double; C-J = core-jet; T = triple; CPLX = complex. ND = Not Detected). Columns 7 and 9 report the total cleaned flux density ad 2.3 and 8.3 GHz respectively, while Cols. 8 and 10 show the resolution achieved in the images at 2.3 and 8.3 GHz. Column 11 indicates where the source has been drawn from: DS is for Dallacasa & Stanghellini (1991), W is for Wills (1979), BMSL is for Barthel et al. (1988). Finally, Col. 12 presents our proposed classification as GPS, CSS, candidate GPS (cG) and flat spectrum source (FS)

IAU Name	ID	z	LogP	LAS	Morph.				HPBW_X	from	class.
			W Hz^-1	mas		Jy	mas (PA)	Jy	mas (PA)
(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)
0201+113	Q	3.61	28.4		P	0.56	25.1 10.5 (60)	0.67	9.7 3.3 (53)	DS	cG
0237-027	Q	1.116	27.3		S	0.26	20.9 13.7 (58)	0.80	7.5 4.3 (64)	W	cG
0237-233	Q	2.223	28.5		S?	3.27	49.7 16.9 (37)	1.93	21.0 4.5 (33)	DS	GPS
0500+019	G	0.583	26.9	10	S	1.56	26.2 11.4 (70)	1.36	9.8 3.4 (68)	DS	GPS
0511-220	Q	1.296	27.5		S?-CJ?	0.33	38.2 19.2 (42)	0.86	14.7 5.4 (49)	DS	cG
0743-006	Q	0.994	27.3		P	0.70	26.7 13.0 (-78)	1.71	8.4 4.5 (84)	DS	GPS
0922+005	Q	1.72	27.6		S?-CJ?	0.45	24.4 12.5 (-73)	0.41	10.0 4.5 (-87)	W	FS
0941-080	G	0.228		50	D	0.92	37.9 18.0 (-57)	0.20	8.2 5.4 (-81)	DS	CSS
0941+261	Q	2.906	27.7		ND	ND	---	ND	---	BMSL	CSS
1143-245	Q	1.950	27.9		S	1.11	40.7 20.3 (41)	0.92	16.8 6.0 (39)	DS	GPS
1237-101	Q	0.753	26.9		S	0.82	23.4 14.3 (51)	0.67	7.9 4.2 (47)	DS	GPS
1317-005	Q	0.892	26.8		P	0.18	20.9 13.0 (-54)	0.11	7.1 3.8 (50)	W	CSS
1502+036	Q	0.411	26.3		S	0.51	28.6 11.3 (56)	0.61	13.5 3.2 (46)	W	FS
1518+047	G	1.296	27.5	150	D	1.69	22.1 10.2 (53)	0.40	10.5 3.6 (44)	DS	GPS
1607+268	G	0.473	26.7	60	D	2.39	18.7 9.9 (66)	0.90	9.1 3.7 (48)	DS	GPS
1629+120	Q	1.795	27.7	1150	D	0.29	31.9 10.0 (53)	0.11	10.0 4.1 (44)	BMSL	CSS
1848+283	Q	2.56	28.1		P	0.24	22.1 10.6 (54)	1.44	8.3 3.4 (51)	DS	GPS
2044-027	Q	0.942	27.1	40	S	0.29	28.0 17.9 (62)	0.03	10.0 4.3 (50)	DS	CSS
2126-158	Q	3.268	28.4		P	0.56	33.6 14.7 (42)	0.86	13.0 3.9 (41)	DS	cG
2128+048	G	0.99	27.5	60	D?	2.25	24.1 12.1 (61)	1.28	8.1 3.7 (53)	DS	GPS
2137+209	Q	1.576	27.4	50	S	0.63	24.6 9.8 (52)	0.23	9.6 3.3 (46)	DS	CSS
2210+016	G	...		100	T-CPLX	1.20	23.6 12.3 (60)	0.40	9.7 3.7 (54)	DS	CSS
2223+210	Q	1.959	27.9		S	0.79	22.3 10.0 (57)	1.23	8.8 2.9 (50)	BMSL	CSS
2351-006	Q	0.46	26.0		S	0.29	24.2 12.6 (60)	0.27	8.7 3.9 (65)	W	?

1402-012	Q	2.518	27.8		P	0.49	19.1 11.8 (-83)	0.29	5.2 3.3 (-84)	W	cG
1602+576	Q	2.85	27.8	120	D?	0.57	17.8 10.6 (89)	0.34	4.1 2.4 (90)	BMSL	CSS
1629+680	Q	2.475	27.7	120	C-J	0.79	20.6 11.0 (84)	0.20	5.2 3.5 (70)	BMSL	CSS
1801+010	Q	1.522	27.7		P	1.64	21.6 13.6 (68)	1.05	5.5 3.2 (80)	BMSL	cG

The VLBI data were correlated at the MPIfR VLBI MK3 correlator in Bonn. A-priori calibration, fringe fitting, editing, self-calibration and imaging were then performed by means of the Astronomical Image Processing System (AIPS) developed at the National Radio Astronomy Observatory (NRAO).

For unknown reasons, at X-band the sensitivity of Onsala in session 1 was about a factor of 100 lower than expected and all the baselines to this station had to be discarded.

Amplitudes were calibrated using 0235+164 (session 1) and 1739+522 (session 2), both assumed to be unresolved on our baselines, and with flux densities measured at Effelsberg during the VLBI run. The a-priori amplitude calibration suffered from the lack of the TSYS measurements at Wettzell, while at other sites narrow-band and time-varying radio interferences at S band corrupted the accuracy of the TSYS reported in the observing logs, and reduced the correlated amplitude due to occasional saturation of the detector. The final effect is a reduction in the accuracy of the absolute flux density calibration and in a reduced dynamic range of the images.

Each source was observed for a number of 13-min scans (3-5, depending on the source declination in session 1 and 8-10 in session 2) in a range of hour angle, in order to allow the imaging of simple morphologies.

In session 1 we also observed 1607+268 (CTD 93, with a well known double structure) in order to test the reliability of our results.

Antenna	Diam.	Sens. (Jy)	Sens. (Jy)
	(m)	X-band	S-band
Medicina (I)	32	240	850
Onsala (S)	20	2070	1530
Effelsberg (D)	100	40	250
Noto (I)	32	1800	1450
Wettzell (D )	22	660	880
Matera (I)	22	3050	3330

2. Observations and data reduction