diapositives has become quite popular and does show remarkably good results particularly with 
submarginal aerial photography. 
The ground survey work of the private photogrammetric organization follows more or less 
conventional routines. Field survey parties are sent out to establish adequate control to allow 
good absolute orientation of the stereo models. Usually a minimum of three horizontal points 
and four vertical points are desired for each stereo pair. Unfortunately, because of the very 
great expense of this particular operation relative to other operations in preparing a large 
scale topographic map, too frequently attempts are made to get along with somewhat less control 
than is recommendable, particularly in areas which are relatively inaccessable to ground 
parties. New surveying equipment such as the geodimeter and the tellurometer are rapidly 
becoming commonplace in the United States amongst commercial companies. However, at this 
time, the great predominance of photogrammetric ground survey work depends upon the optical 
theodolite. Many of the smaller companies and even some of the relatively medium size com- 
panies do not have their own ground survey capability. These companies subcontract such work 
to other commercial organizations that specialize in survey work. 
Aerial triangulation and bridging as a standard procedure is relied upon by relatively few of 
the commercial private organizations. There are probably two reasons for this. To properly 
carry out this type of procedure requires a high degree of technical knowledge and competence 
and also fairly expensive equipment. Both these items are in relatively scarce supply. Of 
course it is not an uncommon situation for many of the small companies, when faced with a single 
stereo model which lacks sufficient control, to bridge across this model by establishing control 
from adjacent models surrounding the weak model. This of course might be considered a form 
of bridging of a very simple and relatively reliable technique. The more classic system of 
triangulation, of perhaps as much as five models or more in one strip or a block of strips, is 
limited to those companies which own universal first order stereoplotting instruments. It is 
notable that there is considerable difference between the approach of a private organization to 
the triangulation problem as compared with that of a government agency. This may well be due to 
the fact that normally government mapping agencies work at scales much smaller than the 
scales which the private organizations are apt to conduct their operations. However, it is quite 
obvious that the triangulation techniques utilized by the government use many more man-hours. 
The actual operations may be highly repetitive, painstaking and slow whereas with private or- 
ganizations premium is put upon speed. Adjustment procedures in a government agency are apt 
to be highly analytical and perhaps with much mathematical analysis. The commercial company 
is quite likely to use a very direct simple approach and make adjustments graphically. The 
commercial companies of course use more control and seek to avoid briding over longer uncon- 
trolled areas. Some companies have utilized electronic computer techniques for adjusting their 
triangulation. Such programs are now available from some of the major computer manufacturers. 
Other companies prefer to use the more obvious graphical adjustment for systematic errors of 
triangulation. Unfortunately there are very few test results available in this particular field to 
arrive at any conclusions as to the efficacy and accuracy involved. The actual final test for an 
aerial triangulation in a commercial operation of course is the accuracy of the ultimate map. 
The user of aerial triangulation or bridging techniques is only aware that his operation is suc- 
cessful when he has complied with the accuracy specifications of his client. Of course it is 
necessary to be very cautious if one does not wish to run the risk of having rejected results, 
which of course can be very expensive to the company. The author has found through personal 
observations and experience that a horizontal and vertical bridge of not more than five uncon- 
trolled stereo models in any one strip has proven to be quite adequate and successful. That is, 
supplementary control for photogrammetric compilation can be established in the intervening 
models which will yield very large scale topographic maps, that is 1:1200, without errors ex- 
ceeding the accuracy specifications. 
A procedure which is used by some companies possessing first-order equipment, which is 
more properly aerial triangulation than bridging, is of interest. This is the technique of flying 
the mapping project at two different altitudes; the lower altitude for the actual compilation work 
and the higher altitude for control purposes. In such situation the higher altitude is usually at 
least twice the lower altitude so that one high altitude stereo model covers a minimum of four 
low altitude stereo models. The field survey parties are then instructed to thoroughly control 
10