Recently, Buta etal. (2009) examined the question Do Bars Drive Spiral Density Waves?, an idea supported by theoretical studies and also from a preliminary observational analysis Block etal (2004). They estimated maximum bar strengths Q_b, maximum sp
iral strengths Q_s, and maximum m=2 arm contrasts A_2s for 23 galaxies with deep AAT K_s-band images. These were combined with previously published Q_b and Q_s values for 147 galaxies from the OSUBSGS sample and with the 12 galaxies from Block etal(2004). Weak correlation between Q_b and Q_s was confirmed for the combined sample, whereas the AAT subset alone showed no significant correlations between Q_b and Q_s, nor between Q_b and A_2s. A similar negative result was obtained in Durbala etal. (2009) for 46 galaxies. Based on these studies, the answer to the above question remains uncertain. Here we use a novel approach, and show that although the correlation between the maximum bar and spiral parameters is weak, these parameters do correlate when compared locally. For the OSUBSGS sample a statistically significant correlation is found between the local spiral amplitude, and the forcing due to the bars potential at the same distance, out to 1.6 bar radii (the typical bar perturbation is then of the order of a few percent). Also for the sample of 23 AAT galaxies we find a significant correlation between local parameters out to 1.4 bar radii. Our new results confirm that, at least in a statistical sense, bars do indeed drive spiral density waves.
We investigate the variation of bar strength with central velocity dispersion in a sample of barred spiral galaxies. The bar strength is characterized by $Q_g$, the maximal tangential perturbation associated with the bar, normalized by the mean axisy
mmetric force. It is derived from the galaxy potentials which are obtained using near-infrared images of the galaxies. However, $Q_g$ is sensitive to bulge mass. Hence we also estimated bar strengths from the relative Fourier intensity amplitude ($A_{2}$) of bars in near-infrared images. The central velocity dispersions were obtained from integral field spectroscopy observations of the velocity fields in the centers of these galaxies; it was normalized by the rotation curve amplitude obtained from HI line width for each galaxy. We found a correlation between bar strengths (both $Q_g$ and $A_{2}$) and the normalized central velocity dispersions in our sample. This suggests that bars weaken as their central components become kinematically hotter. This may have important implications for the secular evolution of barred galaxies.
