HEREDITY, VARIATION 109 



By adding together the numbers on both sides of the 

 curve we find that in this particular group of 2600 men 

 none were quite so short as 4 feet 8 inches or quite so tall 

 as 6 feet 8 inches. But in any other group of the same 

 number the extremes might be a little more or less, perhaps 

 a quarter of an inch or rarely a whole inch. We should 

 have to measure a million, or even several millions, to get 

 the average height and the proportionate greatest and least 

 heights ; and even then we should not get near the absolute 

 limits of our race, as we know that at long intervals giants 

 and dwarfs appear, differing by many inches, or even by a 

 foot, from all others living at the time. But, omitting these 

 rare occurrences, the measurements of a few thousand among 

 a fairly mixed population will give us the mean height of 

 the whole, very nearly ; as well as the proportionate numbers 

 of those of particular heights, as, for example, at 5 feet 3 

 inches or 6 feet 3 inches. But even the mean height does 

 not remain the same if the mode of life changes. It is 

 certain that the larger proportion now living in crowded cities 

 than there were a century ago has considerably dwarfed our 

 population. 



We will now give an example of variation in a wild 

 animal in order to show that man and the animals and 

 plants which he has domesticated or cultivated do not 

 differ in this respect from those existing in a state of nature. 



The diagram here given is formed from the measure- 

 ments of six separate portions of twenty male specimens of 

 the Bob -o'- link or Rice -bird (Dolichonyx oryzivorus), very 

 common in North America. All were obtained in the same 

 place on the same day, so that there could be no suspicion 

 of their being in any way selected as especially variable. 

 It is a little larger than our yellow-hammer, and is there- 

 fore of a convenient size to be shown on a diagram of its 

 actual dimensions, thus giving a better notion of the 

 amount of variation of the several parts than if reduced to 

 a smaller scale. 



The vertical lines, numbered at top and bottom, 1-20, 

 show the measurements of the twenty specimens of this 

 bird, and the figures at the sides, 0-5, mark the inches. The 

 specimens are arranged in the order of length of body, 



