longitudinal distance between two points. The 

 methods and techniques used in accomplishing 

 these two additional steps are tiring and lengthy. 



The maximum total error of this photo- 

 graphic method for some positions can be as 

 large as +0.05 to +0.06 centimeters; though af- 

 ter analyzing the sources of error --the grid 

 lines, the glare, the calibrated icnob, and the 

 steel tape --we feel that the results can be appre- 

 ciably improved. 



7. Many improvements can be made on 

 the photographic device and the method of 

 reading the film and computing the longitudinal 

 distances. The greatest improvement would 

 be the use of an automatic film reader that 

 punches the film measurements on IBM cards. 

 The cards are then fed to IBM machines that 

 compute the longitudinal distances automatic- 

 ally. A less costly device could no doubt be 

 made, however, by sacrificing certain less 

 needed advantages. Tables also could be used 

 in the place of the IBM card and machines . 



Thanks are due Clinton E. Atkinson, 

 Chief of the Pacific Salmon Investigations, and 

 Alvin E. Peterson, Project Leader of the Racial 

 Sampling Group for their full cooperation during 

 the time of development of this photographic 

 method. We are particularly indebted to Rich- 

 ard H. vanHaagen for advice on certain aspects 

 of the theory and for the design of the wiring 

 system, to Don Worlund for the derivation of 

 the formulae used in this work, and to Virginia 

 Coleman for the diagrams. We are also grate- 

 ful to the many other members of the staff at 

 the Montlake Laboratory who gave so graciously 

 of their time and help when called upon to do so . 



Our special thanks go to F. Bruce Sanford 

 who suggested the method used to present the 

 theoretical section. 



BIBLIOGRAPHY 



Abel, Charles and Kenneth S. Tydings 



1951. Modern Exakta guide and reference 

 book . Greenberg Corporation, 

 New York. 124 p. 



8. From our research to date, it is con- 

 cluded that (1) the theory set forth in this paper 

 is sound, (2) that the device can be made fully 

 portable by designing the base so that it can be 

 disassembled and by reducing the weight of 

 certain parts of the device, (3) that the efficiency 

 of this photographic method can be improved by 

 the design of a better method of reading the film 

 and by incorporating IBM cards and machines or 

 tables to simplify the computations, (4) that the 

 device in its present form can be called accurate 

 for some purposes but that the present accuracy 

 can be improved, and (5) that when the above 

 improvements are made, this photographic 

 method will be a valuable aid in the racial an- 

 alysis work being carried out by the U. S. Fish 

 and Wildlife Service for the International North 

 Pacific Fisheries Commission. 



ACKNOWLEDGMENTS 



Valuable assistance was given by Dr. W. 

 F. Thompson of the Fisheries Research Insti- 

 tute, University of Washington, regarding the 

 method chosen - that of using two cameras and 

 that of determining the measurements of fish 

 by triangulation . 



Benson, Bernard S . 



1956. Film evaluation- -problems and 



solutions. Benson -Lehner Corpora- 

 tion, Los Angeles 64, Calif. 15 p. 

 (General description of the Benson- 

 Lehner film reader.) 



Edmund Scientific Corporation 



Homemade camera stereo -adapter. 



Edmund Scientific Corporation, 

 Barrington, N.J. 10 p. 



Hershey Manufacturing Company 



Sun -Lite II portable speed flash in- 

 struction manual. Hersey Manufac- 

 turing Co., Chicago 3, Illinois. 16 p. 



Hexcel Products Incorporated 



Advance excerpts from Hexcel tech- 

 nical brochure "D" Hexcel Products 

 Inc., Oakland 8, Calif. 22 p. 

 (Gives general description of Hexcel 

 sandwich -board . ) 



International North Pacific Fisheries Commission 

 1954. Report of first meeting. Feb. 1 to 12 , 

 Washington, D. C. 40 p. 



16 



