In Paper I, we performed spectroscopic observations on 11 confirmed glob- ular clusters （GCs） in M31 with the Xinglong 2.16 m telescope. We mainly focused on the fitting method and the metallicity gradient for the M31 GC sample. Here, we analyze and further discuss the dynamics, metallicity and age, and their distributions, as well as the relationships between these parameters. In our work, eight more con- firmed GCs in the halo of M31 were observed, most of which lack previous spectro- scopic information. These star clusters are located far from the galactic center at a projected radius of ~ 14 to ~ 117 kpc, which is more spatially extended than that in the previous work. Firstly, we measured the Lick absorption-line indices and the ra- dial velocities. Then the ages and metallicity values of [Fe/H] and [a/Fe] were fitted by comparing the observed spectral feature indices and the Single Stellar Population model of Thomas et al. in the Cassisi and Padova stellar evolutionary tracks, respec- tively. Our results show that most of the star clusters in our sample are older than 10 Gyr except B290, which is ~ 5.5 Gyr, and most of them are metal-poor with metallicity [Fe/H] 〈 - 1, suggesting that these clusters were born at the early stage of the galaxy＇s formation. We find that the metallicity gradient for the outer halo clusters with rp 〉 25 kpc may have an insignificant slope of -0.005 ＋ 0.005 dex kpc-1 and if the outliers G001 and H11 are excluded, the slope does not change significantly, with a value of -0.002 ~ 0.003 dex kpc-1. We also find that the metallicity is not a function of age for the GCs with age 〈 7 Gyr, but for the old GCs with age 〉 7 Gyr, there seems to be a trend that the older ones have lower metallicity. Additionally, we plot metallicity distributions with the largest sample of M31 GCs so far and show the bimodality is not significant, and the number of metal-poor and metal-rich groups be- comes comparable. The spatial distributions show that the metal-rich group is more centrally
