Accepted financial theory holds that stock splits provide no wealth benefits to stock-holders. The corporate management view is that stock splits add value by placing the stock in a more liquid price range. Empirical explanations of excess returns near the split rely principally upon an information effect. Other findings are that (1) an unexplained, sustained jump in returns' variance occurs at the split, and (2) there appears to be a coincidental decrease in liquidity, not an increase. Daily returns from CRSP and daily and intradaily returns and daily trading volumes and price change information from trade-by-trade data are used to examine the returns variance increase and any connection it may have with any liquidity change. Binomial probability comparisons of returns' variance measures each side of the split ex-date are used to examine the variance change and liquidity change phenomena. T-tests are also used to examine the mean-variance change and the possible change in several liquidity measures. Linear regression is used to detect impact of the general market variance level, firm-specific variables, and microstructure measures, and liquidity measures upon the returns’ variance change. Findings include: (1) the variance increase is significant and exhibits a firm size effect and is affected by the previous history of splits use and. dividend payout, (2) the increase is primarily related to the price level adjustment and changes in the liquidity measures, (3) a slight change in the demand to hold as measured by the percentage of the firm traded takes place for firms with an increase in variance (4) the bid-ask spread decreases, but increases relative to the new price. Stock splits with increased returns’ variance have significantly different liquidity measures from splits where the variance declined. / Ph. D. / incomplete_metadata