If the unfortunate Pius IX. had applied his remarkable talents and earlier influence to this aim, instead of the mere problematical and hazardous experiments of political reform, there is no question but that posterity would have classed him with more unanimity among the benefactors of our race; and as little, that he would hardly have been ever, as now, among that band of illustrious exiles whom Europe has so lately seen departing from their former seats.

And in regard to equality of populousness as affecting the results of comparisons between cities, it is presumable that physical discomfort and immorality-the most efficient aids of Death-find always scope more ample in proportion as the subjects are more dense.

I now present the Table spoken of, in which the latitudes are given only to the nearest minute:

Table shewing the influence of Latitude upon the rate of Mortality.

It appears from this that a high northern latitude is favorable to a low rate of mortality: that at the mean parallel, the mortality is the greatest; and that then it is liable to a diminution until the tropics are reached. Within the tropics there are no reliable data; but it is generally admitted that the mortality is there greater than in the temperate zone. This is instanced in two cases on the American Continent, viz:

These statements all relate to the northern hemisphere; for similar ones in the southern, there are no reliable data.

As the object of this Table is mainly to shew the regularity of natural laws when viewed in a sufficient generalization, I have designedly introduced into the several groupes, instances where though the limiting parallels are the same, yet the other elements of climate, height above the sea, shelter, aspect, &c., are materially variant. Thus Madrid, for example, a city situated inland on a steppe some 2,000 feet above the level of the ocean, and exposed to chilled and piercing winds, might be expected to have a very different climate from Leghorn or Naples, which are near the level of and tempered by the sea. Yet unfavorable as its position in some regard might seem to be, its mortality is, in point of fact, low compared with the rest of its groupe and corresponds with the results already found for higher latitude with which high elevations are to a certain extent interchangeable. In strict theory, all these elements of level above tide, mean temperature, mean moisture, exposure, &c., &c., find their place in determining the climate and the natural mortality of any locality; and in any popular discussion even of this point specially, it would be necessary at least to have determined the principal isothermal and isohygral lines (under which the mean annual temperature and dew-point are the same) of the globe. But such a discussion would be too extensive for any present scope; and I shall therefore only farther in connection with this point offer some results as to the influence of the Seasons upon the mortality here.

The health-returns furnish the numerical mortality for every month; and upon those numbers I have constructed, for each year, the following Table; giving throughout, the month in which the number of deaths was the greatest, that in which it was the least, and those in which it comes the nearest to the annual mean (or twelfth part) of the whole annual mortality.

From a table so detailed as this, we can say with certainty, that the maximum of mortality occurs in July and August; once only it passed over into September, and once was translated to March. The minimum, though not so well defined, is seen in April and May; and July is the only month which is not found in one year or another to conform to the But if we arrange the number of deaths according to the seasons

mean.

of the year, marked as they are by unequal distribution of temperature and moisture in the atmosphere, we shall find the annual mortality, with but few exceptions, conforming to the indications of these atmospheric affections. The season whose thermometric and hygrometric condition is the most variant from the mean of the year is that of greatest mortality; and that in which the atmosphere approaches most nearly to its normal state (which we may consider as shewn by the annual mean, and still more when that mean is derived from a number of consecutive years) presents the absolute mortality as the lowest. The other seasons correspond with the same indications.

*

As I have not the data for the atmospheric means through the period in question, I can only make comparison with those for the year 1836, which I have already used elsewhere for another purpose. The seasons are the Spring, comprehending the months of March, April and May; the Summer, corresponding to June, July and August; the Autumn, including September, October and November; and the Winter which coincides with December, January and February. By this arrangement, in taking out the number of deaths for any year, that belonging to December is entered under the succeeding year. And as the years are not continuous in this respect, I have found it proper to divide the series into three groupes which are in a certain sense symmetrical. For greater illustration, I have inverted the natural order; so as to present the Summer and Autumn of each year and period, before the Winter and Spring, Table shewing the absolute number of Deaths (from 1 Dec. 1826 and onwards as under) in Baltimore, distributed according to Seasons and the corresponding conditions of the Atmosphere.

The atmospheric means from observations for a longer period, may be stated at 30 in. for the barometer, and 51° and 35° for the thermometer and dewpoint respectively. The determination of this last is, however, more doubtful; since it depends in part upon the indications of the dry and wet thermometers. They may be taken nevertheless as sufficiently accurate for the present purpose; and the Table then evinces, what was said before, that the nearer the atmosphere is to its normal state, the less is the mortality. The periods, then, of extreme variations in pressure, temperature and moisture under a given climate, are those which are attended with the greatest mortality. How far the suddenness of those variations influences the result is a question which principally concerns the physiologist; but I regret, that I have not the means of offering at present some actual observations on this point also.

Such a rule as has been just stated, reposing upon physical causes of great generality, may be expected to govern throughout, and to be

*Second Report on the Manufacture of Iron, etc. p. 139.

manifested as well in respect to the daily variations of the atmosphere as under such as are observed and averaged for periods more extensive. But in regard to this point-the influence of the hours of the day upon mortality-we are here absolutely without observations; while those that have been made elsewhere, as in Brussels and Hamburg, rest upon too few instances to be relied on. It would seem from these last, that the hours between midnight and morning (6 A. M.) witness the greatest number of deaths; and that in general, more deaths occur at night than in the day-time.

Leaving this point for future investigators and better means, I pass on to another upon which we have here some statistics, and which is in itself of great interest-the influence of Race upon mortality. The population of Baltimore as constituted at different epochs, I give in the following Table; accompanied with a statement of the absolute, relative and proportional mortalities of the two races which compose it.

From this it appears that the increase of the black population, has been at less than one-half the annual rate of augmentation of the whites; and that while the absolute mortality between the two epochs corresponds to an annual increase at 1 per cent. for the whites, for the blacks it is represented by an actual decrease, equivalent 8-10 per cent. per annum. The column of relative mortality, shew that the life of the black race in 1840, was as stable as that of the white in 1830, owing (it may be supposed) to a change in physical circumstances operating upon both though not in equal degrees; for the last column gives 3 per cent. nearly, more improvement in the stability of life (or vitality) of the black race than of the white. Finally by a combination of the proportions given, we are authorised to conclude that this stability of life is for the two races at a mean, as 44 is to 36,9 in favor of the white race. How much of this disparity is due to physiological differences, i. e. to race, and how much is to be attributed to caste, i. e. social and personal relations and comforts, it does not form part of my purpose to enquire; even if I had the data upon which to solve the question.

The periods in the above statement are restricted to two epochs, because they are the only two census-years included in the bills of mortality, available. These give the direct enumeration of the different elements of the population; which for any other epochs could only be made out by similar calculations to those employed for the aggregate population—a labor that does not seem to me essential to repeat. The results as obtained will probably command the reliance of a majority of readers, all the more for their having been obtained by direct instead of inferential methods.

Methods of this last kind, however, find proper scope and indeed are

necessarily to be resorted to in illustration of another point connected with the colored population, viz. the influence of Condition (i. e. emancipation or otherwise) upon the mortality of that race. We have registers, distinguishing the annual number of deaths of the slaves and of the emancipated, for the whole period from 1826 to 1848, with the exception of the year 1832. The absence of returns for this year destroys the continuity of the series, and makes necessary that it be broken up into a number of corresponding groupes; which I have done, with as much symmetry as the case allowed, in the following Table, intended chiefly to shew the proportion which the numbers of deaths in the two conditions bear to the aggregate mortality.

The preceding Table shewed an improvement from 1830 to 1840, in the stability of life of the entire black population; this one shews that it has been gradual and progressive from 1826 to the present time, and that it is confined to the free portion of that population. The proportion of slave-deaths in the whole mortality, has been visibly constant.

No statement of the relative mortality of the two classes is given; because it could only be derived from and apply to two years out of the whole series; and the results, as exhibited by the numbers of those years, would be entirely fallacious if drawn into a general conclusion; since the class of slaves is neither progressive from year to year, nor permanent in any one year. A considerable number annually is transferred to other States. It is from them, besides, that the emancipated class which is on the increase, is continually receiving important additions. Such transitions, by which the same individual comes, within two years, to belong and to be counted in both classes, renders of course the number of the living in either at any time, (unless duly corrected) an inaccurate factor in the proportion. These transitions, which we residing here understand very well, will be manifest to others abroad by the following statements from the three last censuses; where the difference in numbers after the usual age of emancipation (which corresponds also to the period of greatest physical activity and value) is very distinctly marked.

The last column shews the number that would probably survive from the average younger to the average older life, according to the Carlisle