eye-work in using the shorter lines. Indeed, I found that readers could read matter printed in lines of 25 millimeters in one downward sweep without any lateral movement of the eyes. With lines 30 millimeters long, the lateral movement was sometimes almost nil, and seemed to be due mainly to habit. In reading such lines in this way the eye's extent of movement is hardly more than one fourth or one fifth the amount needed for the same matter when printed in long lines.
When the shorter lines, generally, more words were read per fixation than with the longer ones. A magazine column having lines 60.5 millimeters long was in one case read at the rate of 3.63 words per fixation, while columns having lines 98 to 121 millimeters long required a fixation for every two words. Lines of a length approximating 60 millimeters are usual in newspapers and in my experiments were read with a minimum of eye-movement. The makers of the modern newspaper have felt the reaction of readers more, perhaps, than have the makers of books. Out of this experience has evolved the present practise of printing newspapers in narrow columns, the line-lengths of which are perhaps as near the optimum as can be determined at present, when we consider that much shorter lines give great inconvenience to the printer.
For books, also, the newspaper line-length is near an optimum so far as ease and speed of reading are the conditions to be considered. In the case of large books where the question becomes one of printing in one or in two columns per page the latter alternative should undoubtedly be chosen. For books of ordinary sizes a somewhat longer line may be used where this will contribute to convenience or beauty; but a book should not be used whose lines are more than 90 millimeters in length, and somewhat shorter lines are generally to be preferred.
One of the great advantages of the shorter lines is that they constantly permit the reader to see in indirect vision what his eye has just passed as well as what is just coming. Though the words of this related matter may not be clearly perceived, they furnish visual clues which keep the reading range further extended at each moment, a most desirable condition for all reading and especially for fast reading or for skimming. With such lines a hurried reader may glance straight down a page with only an occasional short stop, and may yet be sure that he has gathered the gist of everything.
Dr. Dearborn, in experiments made recently at Columbia University, found that the eye makes its longest pause near the beginning of the line, thus permitting a general preliminary survey of the line. A secondary pause of more than average duration is made near the end of the line, perhaps partially in review. He finds that lines of only moderate length facilitate these general surveys better than the