eliminated by using a posteriori conpensation, this demonstrates 
that for elimination of Systematic errors a posteriori error 
compensation is a good method to be adopted. 
2e Comparison on interpolation method by a posteriori com- 
pensation 
In experiments of a posteriori compensation, two kinds of 
interpolation methods are used, one is the method of weighted 
mean values (namely program 1), the comparison of effect on 
these two interpolation methods can be made by average increase 
of rate of multiple :for mean. square error of coordinate re- ^ 
Siduals, as well as the average percentage of eliminating 
systematic errors, the statistical results are shown in 
Table 4. 
From Table 4 it can be seen that effect on eliminating 
Systematic errors by weighted mean value method is worse 
than that by polynomial method. This proves that interpo- 
lation method has definite influence on effect of a posteri- 
ori compensation, among methods of interpolation which is the 
optimum one still remains to be further investigated. 
(f) Practical Materials 
Test results given from test field of Enping of Guandong 
Province and that of HORDORP are Shown in Table 5. 
In Table 6, for these two sets of data accuracy increase of 
rate of multiple after a posteriori compensation are listed 
Statistically. 
From Table 5 and 6, it can be seen that test field data of 
Enping of Guangdong Province after a posteriori compensation 
accuracy in planimetry and height are all improved, and for 
test field data of HORDORF after a posteriori compensation, 
accuracy in height increase, but accuracy in planimetry re- 
mains constant basically. 
Experimental results carried out by practical materials in- 
dicates that the performance of a posteriori compensation is 
quite stable, its effect on compensation of systematic errors 
in height is more significant than in planimetry, this method 
is effective to the inprovement of accuracy in heighte 
- 500--