In this study, the methodology which based on cross sectional volatility of stock rate of return and was carried out by Christie and Huang (1995) with Chang et al. (2000), was implemented to study the existence of herding behavior in Istanbul stock exchange (ISE) National 100 index. The existence of herding behavior was examined in terms of two models as rising and falling days and asymmetrical and nonlinear relationships were put forward. Index and traded stocks as part of an index were handled between 04.01.2000 to 04.01.2010 during realization of analyses and total 2479 return data on daily basis, which were obtained for every stock and index, were analyzed by using regression method. As a result of analyses, it was found that increasing index return rate in rising days of markets increased cross sectional volatility significantly and these findings were supported by other findings, which were obtained for falling days in markets. When the obtained results are evaluated, it can be argued that herding behaviour is valid in Istanbul stock exchange (ISE) National 100 index and a nonlinear relationship exists between them.