The objective of this study was to evaluate the efficiency of spatial stati stical ana lysis in field trialsrnand, particularly, we have demonstrated the benefits of the approach when experimentalrnobservations are spatia lly dependent. We have compared (i) cl assical randomi zed complete blockrnmodel with independent and identica lly distributed errors (RCBDiid); (ii) the most common spatialrnmodels: Exponentia l, Spherical and Gaussian (with and without nugget effect); (i ii) spatial modelsrnof (ii) with and without block effects; (iv) complete random model (CR). The data used in thisrnstudy were obtained from separate trials for Bread wheat and Durum wheat wh ich was coded asrnBW99RVTII (Regional Varity Trial Two of Bread Wheat conducted in the year 1999 Eth.C.) andrnDW99RVTII (Regional Varity Trial Two of Durum Wheat conducted in the year 1999 Eth.C.),rnrespectively. The restricted maximum likelihood (REML) method was used for est imation ofrnspatial covariance parameters. The denominator degree of freedom of F test for treatment effectsrnwere computed using the Kenward-Roger method for a ll models (Kenward and Roger, 1997).rnSemivar iogram were used for the initial estimate of the parameters of the spatial covari ancernstructure. Akaike's Information criterion (AIC) and Corrected Akaike's Information cri terion AICcrnand the Likelihood ratio test were used for model comparison. The result showed that in all of thernfour trial s, the estimated residual from RCBiid model were s ignifi cantly spat ially correlated,rnproviding evidence of field heterogeneity within blocks that cou ld make the RCBiid method lessrnpowerful than a method that incorporated the spati al correlation. Furthermore, the finally se lectedrnspatial model for a ll data set showed that blocking seemed to be unnecessary if these se lectedrnspatial model were used for each of the trial s. The results also showed that the spatial modelsrnprovided a smaller p-value and standard error than the class ical analysis of variance models. Wernhave concluded that randomizat ion and blocking does not completely remove spatial variation.rnHence, we recommended that researchers should take into account spatial variation in fieldrnexperiment in s ituations were spat ial dependence among the observations is sign ificant.rnKey words: Spatial Models; Semivari ogram