4 
the machine. This can be clone in two ways. Either the plotter looks at the model 
or the picture, and locates the prick on the model in the machine, without marking 
it on the diapositive, or else the prick can be transferred to the diapositive in 
advance with special plotting equipment. A combination of the two methods may 
also be considered. 
The mean plotting error, when transferring was visual with normal angle photo 
graphs, was found to be from a dual plotting + 1.00 m 72 points, of which 50% 
were pricked in treetops, were the basis for investigation. This error includes both 
mistakes in the visual transfer and height adjustments. 
The mean point error for the coordination points has been investigated, but the 
m p shows little variation for the series which were not signalled, and the error 
has little influence on the accuracy with which the boundary points can be deter 
mined. 
The majority of glassdiapositives were copied using methods of contrast adjust 
ment (Kel-O-Wat, Log Electronic), but the quality on the whole was found to be 
poorer than that that of the diapositives obtained through normal printing. 
During plotting of the details from the work done on the ground, the plotter has 
only been able to find about half the signals which were identified by all the field 
workers. To plot points where there were no visible signals, visual transferring 
was used. 
The amount of time used for plotting does not show much variation between the 
different series. The total time used for plotting 100 points graphically was about 
2 1 / 2 -3 hours by the transferring of signals which could not be seen. 
For pure graphical map plotting, when the transfer of points is decided in advance 
and the writing of coordinates is cut out, the time used is reduced to about 1 / 2 hour 
for every 100 points. The same would be true of plotting by computation where 
one has electrical coordinate recording on the machine. Transfer work here comes 
in addition. Se below. 
By stereoscopic transfer with special equipment, the pricked marks in the photo 
contacts are transferred to glassdiapositives before these are put into the plotting 
machine. With equipment made by Zeiss-Autograph, an m p = + 10 is possible, 
corresponding with + 0.15 m on the ground in a scale of 1:15.000. The time used 
when plotting with this equipment was 100 points in 1.90 hours. This included 
marking each point in ink (total number of points done was 1705). 
Grouping of material and the classifying of errors. 
The errors are calculated according to numerically determined coordinates. 
Based on the graphical determination of coordinates, the material has been divided 
into the following groups of error: 
Class I — Errors less than 9 meters made during field work and plotting. Based 
on this, m p is calculated for the comparison of methods. 
Class II — Points which probably are not clearly defined on the ground. The 
field worker has presumably not been able to mark the same points 
in each series (curved boundaries).