850 W..A. Norton—Force of Effective Molecular Action. 
of the more tenacious materials—with a corresponding series of 
experimental values of the coefficient of elasticity for bars of cast 
iron, wrought iron, steel and oak; with the following results, For 
a bar of cast iron, experimented on by Captain Rodman (U. 8. 
Army), the correspondence is very close. The greatest ratio of 
error does not exceed ;'5. For five bars of wrought iron taken 
for comparison, the correspondence proves tolerably close up to 
a stress equal to half the tenacity ; but at the higher ratios of 
stress, the coefficient of elasticity diminishes much more rapid- 
ly than the theory calls for. For the cast steel bar taken, the 
stead of »;. Four bars of blister steel examined present a case 
with it by admitting that some m ing causes are in opera- 
tion which tend to produce abnormal deviations from the theo- 
retical results obtained from our formul ow, as a matter of 
fact, such modifying causes are k exist. We have 
dimensions and its forces, under the operation of varying forces 
of stress; that & is liable to variation, and hence that the 
values of #’ for the same values of « may change, and the mo- 
lecular curve shift its position and rise or fall according as & 
increases under the stress or diminishes. As for the actual 
stress are adequate 
forces 0 
to the production of all the deviations, under consideration, 
represent 
stress amounted to a large fraction of the breaking weight, 
which may be reasonably ascribed to a flow of the molecules. 
