66 



SMITHSONIAN MISCELLANEOUS COLLEC TIONS 



VOL. 71 



In so far as avoiding difficulties of construction are concerned, the use of a 

 smaller number of larger secondaries is preferable, but they should be long 

 and narrow, as otherwise the air "resistance on the nearly empty casings will 

 be greater for the same weight of propellant than would be the case if groups 

 of small secondaries, case (a), were used, in as compact an arrangement as 



(a) 



(b) 



Fig. 8. 



possible. It should be explained, also, that if very small secondaries were 

 employed, the metal of the magazines and casings would become a consider- 

 able fraction of the entire weight, as the amount of surface enclosing the 

 ])ropellant would then be a maximum. 



Possibility of employing case (b). — A rough calculation shows at once the 

 possibility of using a comparatively small number of largo secondaries 



