z 2 
d i 
t Q> 
dj> 
• 
Fig. 1. 
Fig. 2. 
Aeronivellement with statoscop 
On p 137 in [4] v Gruber says: If we want to have a model free 
from parallaxes we have to change the ip-tilt for the preceding photo 
graph, as soon as distortion gives a convergence error or if there are 
small errors in the statoscope heights. 
Nothing is said about the influence of the curvature of the earth. 
Bridging with statoscope and horizon photographs 
The absolute tilts from the horizon photographs are inserted in the 
instrument. 
At the end of the treatise (p. 139) v Gruber writes: »Durch die 
absolute Orientierung der Aufnahmen auf Grund der Horizontbilder 
ist nicht nur der Einfluss der Erdkrümmung, sondern auch jeder 
andere systematische Konvergenzfehler von vornherein ausgeschaltet. 
Das einzelne Teilmodell zeigt keine systematischen Verbiegungen 
mehr, ...» 
On the contrary the nadir directions will be parallell, each model 
b 
obtains a divergence error of y = and a corresponding model de 
formation. 
Convergence errors as a result of the curvature of the earth. 
In the littérature we often (see for instance [3]) find the opinion 
that the curvature of the earth causes a convergence error and a defor 
mation of the single model. 
A model free from vertical parallaxes has no deformation in the ideal 
case. The convergence angle between the nadir directions is true and 
thus the stereo model is conform to the object. 
The curvature of the earth has on the whole nothing to do with the 
performance of the relative orientation, thus it cannot cause a con 
vergence error or a model deformation. 
497