In this paper computer simulation is employed to investigaternthe nature of the error sizes and the distribution of sample sizernrequired in a sequential probability ratio test (SPRT). In therntest we can distinguish between two kinds of error probabilities.rnThe first kind are the specified error probabilities which arernusually denoted by a and B, and the second kind are the truernerror probabilities which may be denoted by at and Btâ€¢ Wald hasrnshown that the relation at + Bt ~ a + B holds true. The objectivernof this project is to investigate the relation between at and a,rnand between Bt and B. As sample size required in SPRT is a randomrnvariable, its distribution is also studied. Two probabilityrndistributions: Bernoulli and normal (known variance) are selectedrnfor the study.rnThe study shows that in ,t.he' case 'of norma'l' distribution,rnwhen the parameters under rio:: 5rrd' :~i' ;~re: ,sl~g~~i~ :f*~: :*p,~'tt andrna = B, the estimates of true error, 'probahiIi ties, C',1:'e less 'thanrntheir corresponding specifi~d ~rror i)r('b~hi!iit~es and"that 'theyrnare close to each other. The est:lmat,es ~t ;t' 'a'h? ~t ,~1~9 decr:easernas d = 9 1 - 90 increases. Th~s, lillJ.q~s, t~a,t, tlJ.~ ~9t~a,i ~iflks, {iz:e byrnfar less than the specified value for large d. When a and Barernnot equal, there are times when the estimate of at or Bt exceedsrnits corresponding specified error size as observed. But, stillrnif the parameters under HO and HI are far apart, the estimatesrnindicate that at ~ a and Bt ~ B.rnFor Bernoulli distribution, the results are not very farrnfrom those of normal except that in some cases the estimates ofrnat or Bt are found to be greater than a or B which led torndisobeying the inequality at + Bt ~ a + B. This may be attributedrnto sample fluctuation.rnAnd finally, sample size distribution is observed to dependrnmainly on d = 9 1 - 9 0 (90 < 9 1). The mean and variance of thernrequired sample size increase rapidly as d decreases. Further,rnthe distributions are all positively skewed. It is also observedrnthat in rare occasions there is a chance that the sample size inrnSPRT exceeds the size one needs in non-sequential test