Development of Form in Literatni'c. ly 



periods and high averages take their place. ^ Instead of a 

 lesser final aggregate for the History on account of the abun- 

 dant dialogue, this was larger than for the Essays by a 

 respectable fraction. The evidence seemed to indicate the 

 operation of some kind of sentence-sense, some conception or 

 ideal of form which, if it could have its will, would reduce all 

 sentences to procrustean regularity. A single act may or 

 may not signify with respect to character, but the sum of a 

 man's deeds for a day or a week will exhibit his ideals and 

 principles and other springs of action. Here, then, in this 

 23.43 was the resultant of the forces which had made Macau- 

 lay's literary character. How the many short sentences are 

 kept at equilibrium by the few long periods is illustrated on 

 next page by a diagram of the sentence-lengths from the 

 first two columns on page 4. The horizontal numberings 

 indicate the sentences in order from one to one hundred ; the 

 vertical show the number of words in the respective periods. 



^ In the History was observed the same fondness for seven as a final digit as 

 had appeared in the figures from the Essay above. There was relatively a great 

 number of sentences — and in one case no less than four consecutively — contain- 

 ing just seven words. Thinking this might be connected in some way with the 

 fact that Macaulay's sentence average was an odd number, I went through forty 

 thousand of the sentences, to ascertain whether even or odd numbers predominated. 

 But I found that the sentences containing each an odd number of words were not 

 more numerous than those of even, as the following summary will show : — 



In first 5000 sentences 2455 even, 2545 odd. 



" second " " 2536 " 2464 " 



" third " " 2462 " 2538 " 



« fourth " " 2482 " 2518 " 



« fifth « " 2491 " 2509 " 



" sixth " " 2504 " 2496 " 



" seventh" " 2537 " 2463 " 



" eighth " « 2534 " 2466 " ■ 



In 40,000 sentences 20,001 even, 19,999 odd. 



But why should the even and the odd sentences alternate in preponderance? 

 This surely could not be fortuitous merely. Other mysteries there were in plenty 

 and seemingly more solvable. The lists abounded in strange runs and ranges of 

 figures, in which it seemed some law should be at once discerned by the mathe- 

 matically or psychologically expert. For my own part, after a few ineffectual 

 attempts to decipher something, I gave up the task. 



353 



