1 82 PSYCHOLOGY AND PREACHING
variation enter in such as individual constitution, the degree of mental training, mental maturity, fatigue, etc. that it is impossible to make any statement about the length of these that is at once general and accurate. Hardly any two persons can be supposed to have an equal capacity to maintain their attention at a given level for a given time. Nor is the same person s capacity hardly ever the same at two different times or with respect to two different objects. All that can be said, therefore, is that there are fluctuations, varying in length with different persons and with the same person under different conditions, which usually last for some seconds, or some minutes; and others longer, which can only be measured by hour periods. It is also well estab lished not only by common experience but also by systematic experiment, that there is a diurnal fluctuation. x " The periods of the day most favourable for work are the former half of the morning and the latter half of the afternoon. The morning period again is better than the afternoon period." This conclusion may seem to be negatived by the fact that many persons find they can do their best work at night. But this is probably due to the fact that they can then work more free from the distractions which fill the hours of the day, and which are so likely to have their effect, no matter how one may strive to isolate himself or how apparently unconscious of them he may be, and not to an increase of the attentive power at that period of the day. However, allowance should always be made for individual peculiarities.
The cause of these fluctuations, it can hardly be doubted, is fatigue. As before noted, it has been suggested that the rapid shifting of the attention is due to the exhaustion of the cells of the brain involved in attending to successive objects. That would imply, however, that each act of at tention called into play a different group of cells from that engaged in the preceding act. Whether that is so or not cannot be determined, and need not concern us here. But
1 Arnold, "Attention and Interest," pp. 91, 247.
�� �
