been slow and gradual. Upon several occasions its boundaries have been accurately defined by our Board of Health. This was remarkably the case in New York, in 1805. The disease in that year was confined for some weeks to a small portion of the eastern side of the city, and, as stated by the Board of Health, not a case occurred in any part of the town, that was not referable to that as its source. In a short time, the infection extended a few streets further; the Board of Health again defined its limits, and again declared that still not a case had occurred, that could not be traced to this part of the city as its source."1 The disease at Gibraltar is almost always confined to the western face of the rock, and to a small village situated at the base of the rock, on its eastern side.

SEC. II.-Season. The period of the year during which yellow fever prevails most extensively, varies with the climate and temperature of different localities. In the cities of the United States, it usually commences in the months of July or August, and continues till the first frost. The great epidemic of 1793, in Philadelphia, began early in August, and ceased about the middle of October; the largest daily mortality taking place during the second week of the latter month. At Seville, in 1800, the epidemic com

1 Hosack's Med. Essays, vol. i. p. 309.

2 [The great epidemic of 1798, in New York, did not commence till about the 14th of August, and it continued till the middle of November. The deaths were two thousand and eighty-six. In 1803, the disease began July 18, and ceased at the end of October. Deaths, six to seven hundred. In 1805 it was declared epidemic, soon after the 1st of September, though scattering cases had occurred from early in June; it ceased late in October. Deaths, two hundred and seventy to three hundred. In 1806 there were a few sporadic cases from June to November. In 1819 it began in the latter part of August, and there were no more than thirty-seven deaths. In 1822 it began on the 11th of July, and ceased early in October. Deaths, one hundred and sixty-six. In 1801 it began as late as the middle of September, and continued till near the end of October, causing one hundred and forty deaths. The disease in New York, when dependent on epidemic influences, does not seem to have ceased in more than one instance (1822) before the middle of October. It has usually continued, but with diminished severity, till about the 1st of November. It has begun as early as the first days of July in one case, as late as the middle of September in one. In 1822 it ceased before the first frost, and in 1800 it continued ten days or more after the first frost. As the average cessation has been the 1st of November, and frost is usually felt in the first fifteen days of October, it is probable that the cold, in the language of Dr. Walters, elsewhere quoted, destroys "the out-of-door poison," but not immediately, "the infection remaining in the houses," An analysis of Dr. La Roche's history of the fifteen epidemics, which have visited

disease of warm climates, and that it prevails most extensively during the warmest seasons of the year, no one pretends to deny or to doubt. But some observers have gone further than this, and have alleged that the disease is much more likely to occur, in the localities that are subject to it, in very warm and wet seasons, than those that are somewhat cooler and drier. They assert that there is a general connection between certain appreciable states of the weather and the disease. Dr. Hosack says yellow fever prevails most extensively when the air is unusually moist and the weather hot. Dr. Doughty says it is more likely to occur in the West Indies, after copious rains. Sir Gilbert Blane said that the fever was restricted to those regions where the range of the temperature was as high as 80°. Others have alleged that the disease can occur only in those places where the average temperature at 3 o'clock P. M. is not less than 79°, during the summer, and especially during the two whole months of June and July; and that its extent and severity will be in proportion to the degree in which it exceeds this point. There is no doubt, whatever, that the disease is generally found where these high temperatures prevail; but it is far from being settled that the disease is directly and immediately dependent upon these degrees of heat; yellow fever has sometimes occurred at Stony Hill, in Jamaica, thirteen hundred feet above the level of the sea, with a mean annual temperature of only 70°.

Sir James Fellowes gives tables of the temperature, furnished by Arejula, at Cadiz, from 1789 to 1803; from which it is quite clear that the hottest years were not the sickliest. They show no apparent connection between the temperature and the disease.' Hillary, who studied this subject with great care, says: "It does not appear, from the most accurate observations of the variations of the weather, or any difference of the seasons which I have been able to make for several years past, that this fever is any way caused or much influenced by them, for I have seen it at all times, and in all ne of the year, in the coolest as well as in the hottest time of it that I have always observed that the symptoms re acute, and the fever usually higher, in a very lly if it was preceded by warm, moist weather,

en it is more cool."3 M. Catel believes that

e of yellow fever at Martinique and at other

menced on the 23d of August, and continued till December; the principal mortality was in October.1

The editors of the New Orleans Medical Journal have published a tabular statement of the cases of yellow fever received into the Charity Hospital of that city, for twenty-one successive years, from 1822 to 1843, inclusive, with the dates of the first and last case for each year. During fifteen of these twenty-one years, the disease was sufficiently extensive to be called epidemic; the number of cases received ranging from ninety-four to eleven hundred and thirteen. The dates of the first receptions vary from May 23 to September 3, the largest number falling in the months of July and August. The dates of the last receptions vary from November 13 to December 31; the largest number falling in the month of November, and almost always after its middle period. Sir James Fellowes has published a similar abstract, in connection with the general Spanish epidemic of 1804. He gives the population of twenty-three towns in which the fever prevailed; the period of its commencement and cessation in each; the day of the largest mortality; the total number of deaths; and the proportion of males and females. The earliest period of the commencement of the disease was June 29, at Malaga; the latest period was October 5, at Villamartin, a small town in the province of Seville; the disease. began in ten towns in August, in nine in September, and in three. in October. The earliest period of the cessation of the disease was October 28, at Grenada; the latest period was January 23, at Carthagena. The disease ceased in the course of October in one town, in November in seven, in December in twelve, and in January in two. Mr. Doughty says that, in Jamaica, the disease generally prevails from the beginning of August to the end of December or January.

SEC. III.-Temperature, and Weather. That yellow fever is a

Philadelphia, shows that in 1699, the disease ceased near the 22d of September; in 1853, on the 7th of October; and in 1747, it "was almost, if not quite gone," on the 16th of October. That in 1820 it continued, with small mortality, till the last of November; and that in all other instances, it ceased with the cool weather of autumn, from the middle of October to the first week in November. The duration of the epidemics varied from two and a half to four months. One or two then ceased before the usual season of frost. Light frosts did not always destroy the poison, and it rarely ceased to operate immediately after a heavy frost. These facts correspond strikingly with those noticed in New York. September has usually been the most fatal month, both in New York and Philadelphia.]

Reports, etc., by Sir J. Fellowes, p. 421.

N. O. Med. Journ., vol. i. p. 103. Reports, etc., by Sir J. Fellowes, p. 478.

disease of warm climates, and that it prevails most extensively during the warmest seasons of the year, no one pretends to deny or to doubt. But some observers have gone further than this, and have alleged that the disease is much more likely to occur, in the localities that are subject to it, in very warm and wet seasons, than those that are somewhat cooler and drier. They assert that there is a general connection between certain appreciable states of the weather and the disease. Dr. Hosack says yellow fever prevails most extensively when the air is unusually moist and the weather hot. Dr. Doughty says it is more likely to occur in the West Indies, after copious rains. Sir Gilbert Blane said that the fever was restricted to those regions where the range of the temperature was as high as 80°. Others have alleged that the disease can occur only in those places where the average temperature at 3 o'clock P. M. is not less than 79°, during the summer, and especially during the two whole months of June and July; and that its extent and severity will be in proportion to the degree in which it exceeds this point. There is no doubt, whatever, that the disease is generally found where these high temperatures prevail; but it is far from being settled that the disease is directly and immediately dependent upon. these degrees of heat; yellow fever has sometimes occurred at Stony Hill, in Jamaica, thirteen hundred feet above the level of the sea, with a mean annual temperature of only 70°.

Sir James Fellowes gives tables of the temperature, furnished by Arejula, at Cadiz, from 1789 to 1803; from which it is quite clear that the hottest years were not the sickliest. They show no apparent connection between the temperature and the disease.' Hillary, who studied this subject with great care, says: "It does not appear, from the most accurate observations of the variations of the weather, or any difference of the seasons which I have been able to make for several years past, that this fever is any way caused or much influenced by them, for I have seen it at all times, and in all seasons of the year, in the coolest as well as in the hottest time of the year; except that I have always observed that the symptoms are generally more acute, and the fever usually higher, in a very hot season, especially if it was preceded by warm, moist weather, than it usually is when it is more cool." M. Catel believes that the epidemic prevalence of yellow fever at Martinique and at other places is greatly favored by the warm and humid winds from the

Hosack's Med. Essays, vol. i. p. 305.

2 Reports, etc., by Sir J. Fellowes, p. 413, et seq.

Rush's Hillary, p. 107.