The total power output of a unit may be represented by the power output of the most productive tube or tubes multiplied by a quantity designated as the ``number of effective tubes.`` The number of effective tubes observed at 105-D under two different arrangements of poisoned columns has been compared with values of this quantity calculated from theory. It was found that in each case the observed number of effective tubes was about 5% less than the ideal number obtainable by an optimum arrangement of the poison. In the case in which two poisoned columns in the pile took up 110 ih it was found that the loss consisted of 2% due to the necessary presence of a control rod in the pile and 3% due to the fact that the actual arrangement of the poison was not the optimum one. In the case in which a single, off center, poison column took up 60 ih, it was found that very little of the loss was due to the presence of the control rod and almost all of it was due to the eccentric position of the P-column. As a further result of this study, it was found that, apart …