Stock Prices and Volume dence of monthly and daily seasonals in means and standard devia- tions of returns. In order to adjust for these documented shifts in both the mean and variance of the price and volume series,we perform a two-stage adjustment process in which systematic effects are first removed from the mean and then from the variance.We use the following set of dummy and time-trend variables in the adjustment regressions to capture these systematic effects: 1.Day-of-the-week dummies (one for each day,Tuesday through Saturday). 2.Dummy variables for each number of nontrading days preceding the current trading day (dummies for each of 1,2,3,and 4 nontrading days since the preceding trading day).These"gap"variables capture the effects of holidays and weekends.The distribution of these gaps in the trading record are as follows: gap of one nontrading day:1339 gap of two nontrading days:1873 gap of three nontrading days:223 gap of four nontrading days:5 3.Dummy variables for months of March,April,May,June,July, August,September,October,and November. 4.Dummy variables for each week of December and January.These variables are designed to accommodate the well-known "January' effect in both the mean and variance of prices and volume. 5.Dummy variables for each year,1941 to 1945. 6.t,t2,time trend variables.(Note:these variables are not included in the mean regressions for the price change.) This list of variables is generally self-explanatory,though we should elaborate on a few points concerning the "gap"variables in the sec- ond group.As to frequency,there are more one-day gaps and fewer two-day gaps than one might expect as a result of trading on Saturdays for the years 1928 through mid-year 1949.As to encoding,if trading occurred on the preceding day,then there is no gap in the trading record and no dummy is included;there are 12,686 such days.The Bank Holiday of 1933 is associated with a gap of 11 days over which the increase in the raw S&P index is the largest close-to-close move- ment in the entire data set.No dummy is included for this single 11- day gap because doing so would,in effect,replace the largest upward change in the price index with the unconditional mean of the price changes,which in our view would not accurately reflect what tran- spired over the Bank Holiday.Finally,in both the adjustments for the mean and variance of volume,the coefficients of the four gap variables 205