In 2004, Yang et al. proposed an efficient (t, n) threshold proxy signature scheme with known signers. However, Maimani et al. showed that a malicious original signer can forge a valid proxy signature for any message in Yang et al.’s scheme and further proposed an improvement to remedy such an attack. In this paper, we will show that in Maimani et al.s improvement, a malicious original or proxy signer can forge a valid threshold proxy signature for any message by different ways. Furthermore, we propose a new threshold proxy signature scheme that remedies the weakness of Maimani et al.’s scheme.