the annual births for eighteen years succeeding this great mortality divided by the annual marriages during that period, give a far greater degree of prolificness than that of the eighteen years preceding that event, (a period which reaches to the commencement of the table,) when similarly calculated. Before the plague, the prolificness of marriages was, on the average, as 3.98 to 1; after that event, as 4.56 to 1: how greatly would the latter proportion appear augmented as compared with the former, had it been calculated on the number of the existing, instead of the annual marriages, at each period! It is clear from the other proportions that the population had not fully recovered itself till nearly half a century afterwards. So strikingly confirmatory are the Prussian registers of the fact pointed out by Sussmilch, that prolificness increases with any great or sudden diminution in the inhabitants of any particular country or district.

(13) Perhaps the plague which raged at Marseilles and the neighbourhood, early in the last century, (of which M. Bertrand has written so affecting an account,) hardly yielded in its severity to that of Prussia; and the consequences, in reference to the subject before us, were precisely the same as those already mentioned in regard to the latter: M. Moheau, like Sussmilch, calls the attention of his readers to the prolificness of marriages which immediately followed that calamity in the south of France.

(14) Nor has the observation of this fact been confined to the old world; the new, again, affords its unequivocal testimony to the true principle of human increase. Dr. Seybert has recorded that the same consequence succeeded the yellow fever in America, which, it has been shewn, followed the plagues of Europe; and has, by anticipation, met the only

VOL. II.

2 N

objection that could be urged against the deduction I have drawn from it. He says, "in Boston, New York, Philadelphia, and Baltimore, where the yellow fever "has occasionally prevailed, there has been an annual "increase in the inhabitants." "We cannot," he adds, "attribute this increase, as they do in other "countries, to the marriages that were influenced by, "or succeeded to, the mortality'."

(15) But this branch of the argument is too important, in all respects, to rest upon general observations and deductions. I shall, therefore, proceed to give it a more particular examination, appealing in its behalf to those documents which are already before the public.

(16) In the numerous tables of Sussmilch, so often appealed to, the epidemic years are marked by him with an asterisk, these (exclusive of those already attentively examined, and also of those found in the sections which are incomplete, the years not being consecutive, and consequently, useless for the present purpose) give the following results. The mortal years so distinguished, collected from the registers of the different countries, which he has published, are 116; the mortality in these years I make to amount to 1,041,346 souls. The same number of the most healthy years, selected, one by one, from every period from which the mortal ones are taken; or, where the latter are placed alone, which is sometimes the case, then from the most healthy year in the two preceding and two succeeding years, gives a total of deaths amounting to 770,054; an astonishing difference, amounting to 35 per cent., of which, thank GOD, we have no experience, and little idea, in this better peopled country. Contemplating, then, the

1 Dr. Seybert, Statistical Annals, p. 49.

dreadful havoc made among the standing marriages, or, in other words, the breeders, by this great excess of mortality, ought not the succeeding year to have exhibited, at least, a proportionate decrease in the number of births, on any other principle than that now advanced? The births, however, of the average years amounted to 1,053,605; and those in the years succeeding those great mortalities, to 1,032,090. The marriages, meantime, in the mortal years, instead of being increased, as Mr. Malthus supposes, were fewer by 2.305. Calculating the fecundity of the two classes of years, by dividing the amount of the births in each by the number of their respective marriages, we find that the quotient is almost precisely the same, namely, 3.7 to each. The only, and inevitable, conclusion, therefore, is, that the fecundity of the remaining breeders had been greatly increased, to keep up the same proportion of births, though the number of existing marriages was so vastly diminished1.

(17) But it has been seen, in every preceding stage of the argument, that the law of population does not manifest itself in the extremes of the argument only, but indicates its existence by a series of intermediate results, graduated, if I may so speak, in conformity to the general principle, with the utmost regularity and precision. I shall, therefore, extend the present inquiry to deviations in the law of mortality, of a less striking character, confined within narrower limits than those hitherto instanced, in fact to such as are perpetually occurring; in order to determine whether the principle which forms the proof now particularly insisted on, is also in minute and constant operation.

See the tables at the end of M, Sussmilch's volume, passim.

548

CHAPTER XX.

OF THE LAW OF POPULATION; AS PROVED BY THE EFFECT UPON HUMAN PROLIFICNESS OF GENERAL VARIATIONS

IN THE LAW OF MORTALITY.

(1) HAVING shewn that the effects of fatal epidemics in any place or country, on the prolificness of the survivors, is in striking conformity to the principle of population, we shall now proceed to examine whether that principle does not receive a further confirmation in the influence which those slighter variations in the annual mortality of a population have upon the conceptions of the periods in which they occur. The very nature of the argument, indeed, demands that the effect should, in these latter cases, be less conspicuous; and it may likewise be rendered still more uncertain or obscure by those extraneous circumstances which will ever have some effect upon the movements of population; still, however, as such circumstances, in any considerable number of instances, may counterbalance each other, we may reasonably expect some indication of the principle at issue, in the facts about to be considered, if it is in reality a law of Nature.

(2) Let us recur, in the first place, to the London Bills of Mortality, in order to determine this matter; and to avoid the possibility of undue selection, take the thirteen years which Graunt has particularized as sickly ones. Such," he says, were

1618, 20, 23, 24; 1632, 33, 34; 1649, 52, 54, "56, 58, and 61, as may be seen from the tables." He adds, afterward, that "the more sickly the years

are, the less fecund or fruitful of children also they be1." If, however, he had recollected that the fairest method of determining upon the comparative fruitfulness of those years, is by the births of the immediately succeeding ones, he would have arrived at a very different conclusion. The actual

deaths in those years amounted to 152,722; the average number, calculated upon his own octennial divisions of the term, to 142,838 only. Computed again on the average of those divisions, the births were 101,359; but the conceptions of the mortal years, given by our authority, amounted to 102,499. Now, to attribute this increase, under such circumstances, to affluxes from the country to the metropolis, in consequence of the room made by this excess of mortality, as he attempts to do when reasoning on the consequences of the plagues in London, were in this case, at all events, too absurd a supposition to be entertained for a moment. It would imply that new-comers took the room which a certain class of theorists assert deaths make for marriages, by anticipation.

(3) Before I conclude my references to this author, I shall again appeal to the registers of the three places which he has given in his work so often alluded to; those of a Country Parish, of Cranbrook, and of Tiverton, going through the several decades into which he has divided them, with the exception of one or two of the last, which reach into the time of the grand rebellion, in which the marriages were no longer celebrated by ecclesiastics nor regularly registered. The following are the results collected, in the same manner as those from Sussmilch's tables. In the most mortal years in

Graunt, Observations on the Bills

of Mortality, c. vi., § 2 and 3, p. 21.

Ibid., Appendix, pp. 47, 48, 49, 50, 51,52, and 53.