SETTING UP OF AN EXPERIMENTAL POLYGON IN ITALY, ECC. 
9 
Ann. N. 6 shows how control points are distributed ; this distribution, on the 
whole, did not differ very much from the net of quadrangular meshes planned. The 
points are numbered from i to 306 : in five cases where two points very near to 
each other were determined they were given the same reference number, preceded 
by a 1 or a 2 respectively : this was done when there were some doubts as to 
whether the marking would be visible in the photograph. 
For each point determined we made out a monography (see ann. 2) divided 
up into two parts, the one containing a description of how the point had been 
pegged out and its location, and the other the numerical elements determined. The 
monography should possibly be completed with a photograph, taken at the stereo 
comparator, of the portion of diapositive showing the point in question. The mono 
graphic parts of the points and the numerical parts of some of them, were to be 
used by the operators to effect bridgings and calculations ; the numerical parts 
of the other points will instead be used to check the accuracy arrived at in the results. 
4) Calculation operations. 
At the beginning of July the measurement operations for the basic net were 
terminated and we proceeded with adjustement calculations, in order to calculate 
all other points as soon as measurements had been completed. 
The planimetric calculations were performed on the plane of the Gauss pro 
jection, referred to the international ellipsoid on the zone of 6° width, with the 
central meridian corresponding to 9 0 longitude east of Greenwich (zone N. 32 
according to the international code) adopting for the E coordinates a false origin 
of 1500 km being, in fact, the reference usually adopted for the coordinates of 
trigonometric points in Italy. 
For the basic net the planimetric and altimetric adjustments were effected 
separately. The first was carried out by adjusting all points in a single block, 
using an IBM 704 electronic computer, according to a calculation program laid 
down by the writer. 
The program allows adjustment by variation of coordinates, on the plane of 
the Gauss projection, of a mixed triangulation and trilateration net including up to 
100 points, up to60 of which to be adjusted, each station comprising up to 12 observed 
directions. The program further allows to introduce into the calculation up to 150 meas 
ured distances. Input elements for this calculation are the coordinates of the points, 
which are considered as provisional values for the points to be adjusted, the ob 
served directions and the measured distances. 
The calculation is effected in one elaboration only, which has to correct the 
ellipsoidal elements to reduce them to the projection plane, to form and solve the 
normal system, to calculate the adjustment corrections of coordinates, directions and 
distances, to calculate the orientation constants of the trig stations, and the for 
mation of the adjusted grid azimuths, to calculate the plane and ellipsoidal lenghts 
of sides, to calculate the mean square error of the weight unit and the mean