84 MAN THE ANIMAL 



to know exactly how many human beings are or were living on 

 the face of the earth at any stated moment. The primary diffi- 

 culty arises from the fact that the process of census counting has 

 never included all the people or areas of the globe j and a 

 secondary one from the fact that in different countries censuses 

 are taken on different dates and with widely differing degrees 

 of accuracy. But the situation grows steadily more encouraging. 

 At the present time it can be said that something of the order of 

 75 per cent of the present population of the world is either 

 regularly counted in recurring censuses, or has been counted at 

 some time in the past, or at least an attempt has been made at a 

 count. Of the wholly uncounted residue, for which there have 

 never been many figures except estimates as a base, China consti- 

 tutes by far the major portion. According to what seems to be 

 the most reliable estimate, China's population at the present time 

 constitutes something of the order of 22 per cent of the world 

 total. When China takes a real census, if she ever does, the 

 study of world population will then rest upon a basis of suf- 

 ficient statistical security to justify a very substantial degree of 

 confidence in results and inferences. 



It is the practice of our laboratory to make from time to time 

 revisions of the figures for world population. The plan used is 

 to assemble separately on work sheets the latest available data 

 regarding population, area, and time (date of count or estimate) 

 for each of the smallest areal and political divisions of the earth 

 for which data are available, and then combine them into larger 

 divisions by summation, checking the figures at each stage in 

 every way possible. 



The data of three recent revisions are given in Table 3. 



The steady decline in the area figures between the successive 

 revisions appears at first glance completely paradoxical, not to 

 say verging slightly on the idiotic. The earth's surface is plainly 

 not shrinking at any such rate. The land area of the globe 

 apparently shrivelled, according to this table some 717,212 

 square miles between average dates of 1930.1 and 1934.7. All 

 that this means is that two things have happened. The first is 



