July 12, 1888] 



NATURE 



247 



'. was 44,107,869. In the Census Report it is, however, 

 shown that over a million females between the ages of 5 

 and 20 must have escaped enumeration ; and when 

 allowance is made for them, the probable accurate total 

 comes out 45,232,391. During January and the first 

 seventeen days of February the increase was 118,532; 

 so at the beginning of 1881 the population stood at 

 45,113,859. From this starting-point the following figures 

 have been worked out : — 



Number living 

 at Commencement. 

 46,794,604 

 47.343.463 

 47.930,770 

 48,241,730 

 48,480,586 



y Number living 



at C immencement. 



1878 ... 45,890,988 



1879 ... 45,628,357 



1880 ... 44,879,553 



1881 ... 45,113,859 



1882 ... 45,688,103 



1883 ... 45,980,913 



Year. 



1884 



1885 

 1886 

 1887 

 1888 



The mean number living during the ten years was 

 46,478,714. 



The total area of the united provinces is given in the 

 Census Report as 106,104 square miles. The population 

 is thus at the present time about 457 to the square mile, 

 including in the average the Himalayan province of 

 Kumaon, over 12,000 square miles in area, where 

 the average density is less than 90 to the square mile. 

 There is practically no export trade, except in agricultural 

 produce ; hence the whole population is supported directly 

 or indirectly by the agriculture of the province ; and there 

 is probably no purely agricultural country in the world, 

 except perhaps some parts of China, where so dense a 

 population is maintained. 



The birth- and death-rates and rate of increase or 

 decrease each year, calculated on the usual basis of iooo 

 livinsr, are given in the next table. 



Mean 



45-40 



39 91 



5-49 



The birth-rate, even in the worst years of the ten, was 

 as high as in England, while in the best years it was 

 about 50 per cent, higher. The death-rate averaged 

 nearly forty per mi He, and therefore, notwithstanding the 

 high birth-rate, the population increased only at the rate 

 of 5'5 per thousand per annum. 



A glance at the annexed diagram will render these data 

 more intelligible. Fig. 1 exhibits the movement of the 

 total population from year to year ; and the nearly straight 

 line, marked Fig. 2, shows what this movement would 

 have been had it proceeded uniformly at the rate of 5-49 

 per mille per annum. The curve has been computed by 

 the formula P = P r„, where n is counted from the 

 beginning of 1883, and consequently the ordinate for 

 1883, P , is the geometrical mean of all the ordinates of 

 Fig. 1. The differences of the ordinates of the first two 

 curves are charted in Fig. 3, which therefore exhibits the 

 extent to which the actual population exceeds or falls 

 short of that given by a uniform movement. This is 

 apparently a periodic function of the time ; and if so, 

 the period does not differ much from ten years, since the 

 last ordinate is only slightly greater than the first. Figs. 

 4 and 5 represent the birth- and death rates respectively. 

 At first sight these appear to have no relation to each 

 other, as concomitant and opposed variations are nearly 

 equal in numbers. The years of lowest death-rate, 1880 

 and 1883, were, however, followed by the years of 



highest birth-rate, showing that healthy conditions con- 

 duce to fecundity as well as diminished mortality. An 

 exception to this rule is, however, found in 1879, which, 

 following a healthy year, should have had a high birth- 

 rate, but was marked instead by a.n exceptionally low 

 one. The year 1878, though a dry and very healthy year, 

 was one in which the vitality of the people reached a 

 low ebb, owing to long-continued scarcity, approaching 

 in some places to famine ; for, though few or none 

 actually died of starvation, millions were for many 

 months at starvation-point. 



The annual rate of increase per thousand, shown in 

 Fig. 6, is probably the best possible measure of the general 

 well-being of the people, combining as it does the effects of 

 abundance or scarcity of food, which influence the birth- 

 rate, with those of health and disease, on which the 

 death-rate depends. Curiously enough, this index of 

 general prosperity or the reverse is much less liable to 

 sudden fluctuations than the birth- or death-rate alone, 

 and yet, like the numbers represented by Fig. 3, it is 

 apparently subject to a periodic oscillation about a mean 

 value. The length of the period is probably something 

 over ten years, since the last year gives a considerably 

 greater result than the first, though it exhibits a down- 

 ward tendency. It is therefore possible that the rate of 

 increase of a primitive people, living a natural life un- 

 trammelled by too much civilization, and multiplying up 

 to the limit of the means of subsistence, may be subject — 

 like the prices of grain, investigated by Mr. E. Chambers 

 and the late Mr. Stanley Jevons, and like many other 

 terrestrial phenomena — to a periodic variation determined 

 by that of the energy received from the sun. Assuming 

 that there is a variation with a period of eleven years, 

 the rates of increase charted in Fig. 6 lead to the 



formula, r = 4-576 + 11-725 sin ( + 262 J° )• This for- 



mula gives the smoothly flowing curve of Fig. 6, which 

 coincides, as fairly as may be, with the curve of actual 

 variations. For the minimum epoch the formula gives 

 1878-73, and for the maximum 1884-23, — dates which fall 

 suggestively near those of the corresponding phases of 

 solar disturbance. 



Into this interesting speculation it is impossible at 

 present to enter further, beyond remarking, as was said 

 at the beginning, that the increase of the population 

 during the last ten years was probably above the average, 

 and too rapid to be maintained. The hypothesis that it 

 is subject to a variation in the eleven-year period leads 

 to the result that the mean for a long term of years 

 is only 4-576 per thousand, instead of 5-49. Now, in the 

 Report on the Census of 1881, the Census Officer, Mr. 

 Edmund White, calculated that the population, as re- 

 ported, increased only 2-33 per thousand per annum 

 between 1853 and 1881 ; but it is pointed out that this 

 result was vitiated by an over-estimation in 1853, when the 

 individual members of the population were not counted 

 by name, but only the total number of each family was 

 entered in the census forms. In the sixteen years from 

 1865 to 1 88 1 the rate of increase was 4-48, and as these 

 years included a fair proportion of good and bad, the 

 rate of movement is probably near the truth. It differs 

 only by a small fraction from the mean rate given by 

 the above formula, according to which the population 

 might be expected to double itself in 152 years, notwith- 

 standing the already great pressure on the soil. In the 

 same Census Report, from the distribution of the popu- 

 lation according to age, a mean death-rate of 39-5 per 

 mille is arrived at. This agrees sufficiently closely with 

 the rate here found to warrant the conclusion that the 

 corrections applied to the numbers actually registered 

 cannot be far wrong. 



Fig. 7 shows the variations of the average rainfall of 

 the province, for which the general mean of the ten years. 



