Hempenius, Makarovié, Van der Weele, Tests of Restitution Instruments 
Those instruments which have in addition the possibility of a graphical presentation 
do not offer any difficulty. The transfer of information from model-space to drawing table 
can be checked separately from the rest of the instrument so that we will not enter into 
this here. 
Fig. 1. Each horizontal plane in this figure contains a set of projected grid points, the 
construction of which is indicated for the vertical plane through the base. The thick lines 
illustrate the extension of the model in space. 
Instruments having only a graphical output can be checked in the same way as 
described in the foregoing sections, the only difference being that the X- and E-coordinates 
of the projected points cannot be read directly but could be measured with a coordinato- 
meter from the plotting sheet. This, however, introduces at least a loss in precision due 
to the intervening medium (sheet) and the additional measurements. 
Since this loss in precision, generally, is not present in the Z coordinates the conclusion 
will be that the residual Z errors and in many cases the residual Y parallaxes are the main 
criterion for the judgement of the geometrical performance of the instrument. Not only 
deficiences in the adjustment of the instrument but also other shortcomings will generally 
be found more easily from Z residuals than from planimetry which, however, does not 
mean that a check on the planimetrie projection should be neglected. Tolerances for residu 
als will, for this purpose, be fixed in terms of units on map scale.