Subband techniques have been recently developed for adaptive filters, since some of the applications such as acoustic echo cancellation and wideband active noise control need adaptive filters with hundreds of taps, which result in high computational complexity and low convergence rate. By the use of subband adaptive algorithms, the computational complexity may be reduced along with convergence rate; however, a delay is introduced in the signal path. To remove the delay, the delayless subband adaptive filter architecture with both the open loop and the closed loop structures have been introduced. This paper presents a new open loop delayless subband adaptive filter structure with critical sampling, where the performance concerning the mean square error of the subband adaptive algorithm, caused due to the aliasing existing in the subband structure, is superior to the results obtained up to now for open loop delayless structures.