We update and extend the results of Shanks ( 1997 ) by making a direct test of Tully-Fisher distance estimates to thirteen spiral galaxies with HST Cepheid distances and to ten spiral galaxies with Type Ia supernova ( SNIa ) distances .
The results show that the Tully-Fisher distance moduli are too short with respect to the Cepheid distances by 0.46 \pm 0.11mag and too short with respect to the SNIa distances by 0.49 \pm 0.18mag .
Combining the HST Cepheid and the best SNIa data suggests that , overall , previous Tully-Fisher distances at v \sim 1000 kms ^ { -1 } were too short by 0.43 \pm 0.09mag , a result which is significant at the 4.6 \sigma level .
These data therefore indicate that previous Tully-Fisher distances should be revised upwards by 22 \pm 5 % implying , for example , a Virgo distance of 19.0 \pm 1.8Mpc .
The value of H _ { 0 } from Tully-Fisher estimates is correspondingly revised downwards from H _ { 0 } =84 \pm 10kms ^ { -1 } Mpc ^ { -1 } to H _ { 0 } =69 \pm 8kms ^ { -1 } Mpc ^ { -1 } .
There is evidence that the Tully-Fisher relation at large distances is affected by Malmquist bias .
In this case , we argue that H _ { 0 } < 50kms ^ { -1 } Mpc ^ { -1 } can not be ruled out by Tully-Fisher considerations .