This Table contains the mortality from fevers of every description among the troops. In Canada, Nova Scotia, Bermuda, Gibraltar, Malta, Cape Town, Mauritius, and the home stations, the form of the disease was almost exclusively continued, with yellow fever at Bermuda, in 1819, and at Gibraltar, in 1828. In the Ionian Islands, the remittent was by far the most fatal form. In Ceylon, the mortality was caused by the remittent form; and in the West Indies, Jamaica, and the Coast of Africa, either by the periodic forms, or yellow fever. The mortality from yellow fever among the population of New Orleans is from Barton.* In addition to the deaths from the yellow form of the disease, however, it would appear, from a Table given by the same author (p. 271), that from 1849 to 1853 a yearly mortality from other forms of fever-4 to 6 per 1,000 of population-occurred; but as it was impossible to obtain this for previous years, the rate for yellow fever alone has been given above. The deaths among the civil population of Malta, from fever, are from data in the Statistical Reports on the Health of the Army; it will be seen that these, embracing, as they do, persons under different circumstances as to age, social position, exposure, and acclimatization, from the troops, do not present an exact correspondence in the increase or diminution of mortality with them, though in the main they agree pretty well. The deaths in the squadron on the Coast of Africa are from the "Report on the Climate and Diseases on the African Station," p. 177. These, as well as the deaths among the military officers on the Coast of Africa, embrace those from all diseases; of the former, it is estimated that all above 20 per 1,000 of mean strength are from fevert, and as 77 per cent. of the latter were from fever also§, (though they cannot be separated for each year), a pretty close estimate may be formed from them of the progress of that form of disease among each class. The rates of mortality from fevers is given for the English troops at St. Helena, from 1818 to 1821 inclusive; and for the civil population, that for 1826-7, and 1831 to 1835 inclusive. This information is less complete than is desirable, but more extended and precise data were not available.

In extended commands, such as the West Indies, Canada, Ionian Islands, and Ceylon, which embrace several subordinate stations, at all of which the maximum or minimum of mortality may not occur in the same year, the aggregate, as given in the Tables, may present neither the sudden increase nor decrease seen at others where the troops were closer together, as at Malta, Gibraltar, or Bermuda. At the latter, too, excessive or very low rates of mortality are more confined to single years, while in the former these are usually more diffused. A large mass of civil population resembles the more extensive military commands in these respects, as may be seen by comparing the deaths among the troops and civil population at Malta.

On examining Table III, it is found that the mortality from fevers undergoes great fluctuations at the same station in different years, and that these fluctuations can be traced through many stations in the same year, or in those immediately preceding or following it, with extraordinary regularity.

Taking Jamaica, the locality of all others in the Table remarkable for its enormous and sustained death rate, the first considerable remission in mortality is found in 1823, when it stood at 52.7 per 1,000 of strength; the next is in 1829, when it fell to 40'6 per 1,000; and the last occurs in 1836, when it was 399. Coincident with the first, very low ratios are found in the West India station, at Gibraltar and Malta; with the second, in the West Indies, Gibraltar, Malta, and Ceylon. The last period, embracing 1830-36, displayed a much less intensity of fever at all the stations in the Table, and the fluctuations of mortality are less marked than in the former two. Each will form a convenient subdivision for more extended examination.

There were, unfortunately, no returns for several of the stations in 1817, and the information for that year in Table III is incomplete. It may be observed, however, that there was a high rate of mortality from fever in the

"Cause and Prevention of Yellow Fever," p. 100; Chart and Comparative Table following Introduction.

+ "Statistical Report on the Health of the Troops serving in the Mediterranean," p. 69, et seq., 1839.

Climate and Diseases of African Stations," p. 178.

§ "Statistical Reports, Western Africa,” p. 24.

Ionian Islands, New Orleans, and Canada, with a moderate one in Ceylon, Malta, and Nova Scotia, while at Bermuda there was none. In the West Indies there was a high mortality this year, and in Jamaica an equally great one, though, when compared with the subsequent ratios for that island, it must be regarded as low.

In 1818 the mortality in the Ionian Islands and Canada fell about one half, and in Nova Scotia and New Orleans there were no deaths from fever. In Ceylon, Malta, and Gibraltar, there was a great increase; in the West Indies the previous year's rate was maintained, while Jamaica and Bermuda both had large accessions. Of the stations south of the line, St. Helena had a high rate of mortality from fever among the white troops, and Mauritius, in nearly the same latitude, a moderate one; while the Cape, intermediate between them in longitude, but more southerly in latitude, had none.*

These facts show that

the influence which led to such high rates of mortality in 1817, in the Ionian Islands, New Orleans, and Canada, was passing off, while another, apparent in the West Indies in 1817, was this year still in operation there, and at St. Helena, and had become highly developed in Ceylon, Malta, and Gibraltar, and was becoming apparent in Jamaica and Bermuda.

In 1819 the death-rate in Ceylon was not more than a third of that of the previous year; it was also much diminished in Malta, Gibraltar, and the West Indies, but it was higher in the Ionian Islands, enormously increased in Jamaica and Bermuda, and there were also high rates at New Orleans and Nova Scotia, the previous year's rate in Canada being maintained. St. Helena had a ratio only one-fourth of that of 1818, and the officers on the Coast of Africa were also healthy, while at the Cape there was a considerable mortality for that station, and at Mauritius a moderate increase. Thus the influence which in 1818 had led to such mortality from fever in Ceylon, St. Helena, Malta, Gibraltar, and the West Indies, was this year passing away to the northward, as indicated by the increased ratios in the Ionian Islands, Jamaica, Bermuda, Nova Scotia, and New Orleans; while the mortality at the Cape, after the freedom from it in the previous year, indicates the advent of a fresh influence there, the progress of which will be traced below.

In 1820 the rate of 1819 continued at the Cape; there was an increase at St. Helena, on the Coast of Africa, and in the West Indies it was very considerable; the previous rates at Malta and Gibraltar were maintained, or slightly increased; those at Jamaica, New Orleans, and Nova Scotia were reduced, but all still high; while at Mauritius, Ceylon, the Ionian Islands, and Bermuda there were great reductions. The influence which appeared at the Cape in 1819 was then apparent this year at St. Helena, the Coast of Africa, and the West Indies, and partially, perhaps, at Gibraltar and Malta, while that which had preceded it was decreasing at Jamaica, Bermuda, and the American stations. Mauritius and Ceylon were little implicated.

In 1821 the Cape and St. Helena were without any death from fever; those on the Coast of Africa were less than a-half of the previous rate, and there was a diminution in the West Indies, though the rate remained high; there was an increase at Mauritius and Ceylon, also in the Ionian Islands, Gibraltar, Bermuda, and Canada, while at Malta the mortality was less both among the troops and civil population. There was a diminution in Jamaica, and no death at New Orleans or Nova Scotia. These facts indicate that the influence which in 1820 was still strongly felt at Jamaica, New Orleans, and Nova Scotia, had this year passed on from there, but was still experienced in Canada; and that that which, in 1819, had manifested itself at the Cape, had now passed to the northward, was becoming weaker on the Coast of Africa and West Indies, but had become apparent at Bermuda, Gibraltar, and the Ionian Islands. There was an increase at Mauritius this year, with a moderate one at Ceylon : this would appear to be unconnected with that influence which has been traced as far as the Ionian Islands, following closely upon it, but extending less to the westward.

In 1822 there was again a considerable mortality at the Cape; there is no

It must be remembered that, as sickness usually occurs as the summer heats decline, the mortality in the southern hemisphere, in any given year, precedes that in the northern by about six months.

record this year for St. Helena. On the coast of Africa the rate was low; in the West Indies considerably reduced; in Jamaica, at New Orleans, and Nova Scotia the death-rate was much increased, while at Gibraltar and Bermuda, and in Canada, there was a great reduction. Mauritius this year had no death from fever, but Ceylon and Malta had higher rates; while in the Ionian Islands that of the previous year was much reduced. The influence there, which in 1821 was in full operation in Canada, was much diminished in 1822, while the following one, which was traced from the Coast of Africa, through the West Indies, to Gibraltar and Bermuda in the former year, had passed away from the Coast of Africa, and was now diminishing at the others, but was in full force in Jamaica, New Orleans, and Nova Scotia. The diminished mortality in the Ionian Islands indicates the cessation of the influence experienced there the previous year; while the absence of mortality at Mauritius, with the increase both at Ceylon and Malta, show the progress to the northward of that noticed last year in connection with the former two stations.

In 1823 the mortality at the Cape underwent little change, but that on the Coast of Africa was much increased. In the West Indies, Jamaica, and Gibraltar, the rates were the lowest since 1817, and at New Orleans there was not a death from yellow fever. At Nova Scotia the rate for the preceding year was maintained, while Bermuda and Canada both had a considerable increase. There was thus one influence approaching from the southward, and another still lingering to the northward, while from Gibraltar to New Orleans, including the West Indies and Jamaica, there was less fever than for years before. To the eastward of the Cape, Mauritius had a slight death rate, that in Ceylon was less; it was also less among the troops in Malta, though it continued high among the civil population, and in the Ionian Islands there was also a considerable increase. That influence, therefore, which was perceptible at Mauritius and Ceylon in 1821, was now diminishing, both at Ceylon and Malta, among the troops, but was operating more strongly in the Ionian Islands.

The facts detailed above, showing the regularity and progression of the oscillations of febrile disease at so many points on the earth's surface, far removed from each other, and embracing the continued, remittent, and yellow forms of the disease, leave no doubt as to the operation of some cause equally extensively. There seem to be a series of waves, generated in southern latitudes, which flow to the north or north-westward in succession, leading to an increase of fever at every point over which they pass; and, in some instances, these are so close, that Canada is under the influence of one, the West Indies of the following one, and a third is apparent at the Cape. The atmospheric vicissitudes, or local influences, to which epidemics are so often referred, confined as they are to limited tracts, are quite insufficient to account for such widely spread effects, though, when the general cause is operating, these local causes are found most influential in generating disease within their sphere of action. A cause, extending over continents and seas as this does, is more correctly designated "Pandemic" than "Epidemic," and as the latter term has come to be applied to every local outbreak, and very often without any attempt to distinguish between an epidemic and a mere increase in the ordinary endemic diseases of the place, it would be well were "Pandemic" to come into use, to indicate that more general influence to which all the others are subordinate. In what follows, the expression "Pandemic Wave" will be employed in the sense here indicated.

The existence and progress of the pandemic wave has been deduced mostly from the records of military stations; but striking evidence of its passage is derived from the outbreaks of disease in the various civil populations within its influence, of which a few instances may be given, as regards febrile diseases, to fill up the sketch. Thus petechial typhus was rife at Naples in 1817, and a disease denominated malignant typhus prevailed at Algiers the same year. In 1818-19 plague ravaged Morocco and the northern coast of Africa, and in 1819-20 it prevailed at Constantinople. Yellow fever prevailed in the south of Spain with varying intensity from 1819 to 1821. There was great mortality from typhus in Ireland in 1817-18-19, and again in 1822. Fever was also very common both in England and Scotland during the former period.

The admissions from continued fevers at the Fever Hospitals of London,

Cork, and Dublin, and the Royal Infirmaries of Glasgow and Edinburgh,

From these it is obvious that the pandemic wave which was still sensible in the
Ionian Islands and Canada in 1817, overspread the British Isles in 1817-18-19;
and it will be observed that London and Cork were, comparatively speaking,
affected to a greater extent in 1817 than Dublin, Glasgow, or Edinburgh; all
except London had their greatest sickness in 1818, but the Scotch stations
carried more of this into the following year than the Irish. The wave described
previously as having passed over the Mediterranean in 1818-19 was experienced
much more slightly; it seems to have kept up the admissions in London during
1819-20; in Dublin, in 1820-21; and in Edinburgh till 1822. The wave
which was at the Ionian Islands in 1821 was sensible in London and Cork in
1822, and in Dublin and Glasgow in 1823. This was followed by that which
was at the Ionian Islands in 1823, so closely, that fevers increased in numbers
for several years, as will be shown below.

Even in Iceland, 1819 was a very sickly year; a fever denominated "bilious”
having prevailed extensively. This was followed in 1820-21 by a "catarrhal
fevert." In 1817 yellow fever was common in the Southern States of America;
there were outbreaks also in 1819-20; and in 1822 it extended as far north as
New York, and even Boston. The coincidence of all these with the passage
of the pandemic waves already described is obvious.

The Island of Ascension occupies rather an important position in the history
of yellow fever, from the supposed importation of the disease by His Majesty's
ship "Bann," in 1823. The island lies nearly 1,000 miles south of Sierra
Leone, in the middle of the Atlantic, and is little else than a mass of volcanic
ashes. The following numbers of its small garrison appear to have been placed
on the sick list for fever, in each period mentioned :-

On the 11th May, 1823, yellow fever broke out, soon after the arrival of
the "Bann." This, in about two months, carried off 15 out of 35 officers and
men, and 6 out of 17 women and children residing on the island. The fatal
case in 1818 presented the characters of yellow fever.§

It has been shown above, that in 1818 St. Helena, which lies 700 miles to
the south-east of Ascension, was under the influence of a pandemic wave, the
operation of which was also distinguishable in the West Indies and Mediter-

The numbers in this and the subsequent Tables for the same places are taken
from "Murchison's Treatise on Fevers," p. 37. The admissions for Edinburgh for
1816 are from Dr. Christison's original paper.

+ Brit. and For. Med. Chir. Rev., Vol. V. p. 464.

Burnett, "Report on the Fever in His Majesty's Ship 'Bann,'" p. 67-9.
§ Ibid. p. 74.

ranean; and the returns for Ascension show its influence to have been experienced there in a greatly increased number of febrile cases, and in the occurrence of a fatal case of the yellow form of the disease. It has been shown, also, that a pandemic wave which appeared at the Cape in 1822, continued in operation there in 1823, but had then extended to the Coast of Africa as well, and consequently embraced Ascension, as on the previous occasion. Under these circumstances the "Bann" arrived, at the commencement of the season which had proved unhealthy in 1818, and which, with a pandemic cause in operation, was that at which disease might have been expected to show itself.* Fever did break out, and the analogy of the occurrences in 1818 (when there was no question of importation to complicate them), is in favour of this having done so in virtue of the action of local causes, under the influence of the pandemic wave.

The circumstances attending the appearance of yellow fever at Ascension, in 1838, a period when there was a strong pandemic cause in operation, not only along the Coast of Africa, but also in the West Indies, are to the same effect. On the 16th and 17th March of that year there were heavy rains at Ascension, which penetrated the houses, leaving a large quantity of mud in several, and formed pools among them. The disease broke out in the end of March, as these dried up. The subject of the first case slept in a room under the surface level of the ground, into which muddy water had found its way during the rains, and the ventilation being bad, some time necessarily elapsed before it could be dried. Vessels had arrived from the Coast of Africa with the disease on board, before it appeared on shore; but it is admitted that the first case was not in a position to have been exposed to personal causes from the vessels, which throws us back on the local causes to account for the production of the disease.

As it appears from the details given above, that in different countries different forms of fever prevailed, under the same general influence, it must be admitted that the pandemic cause determines the frequency and severity, rather than the particular form of the fever, which, there are many reasons to conclude, is more intimately connected with the local circumstances at the time. As bearing on the possibility of yellow fever originating at Ascension, it may be mentioned that at St. Helena a case was returned as this disease in 1819, another in 1820, and a third in 1821; there are no details of these, it is true, but the Medical Officers who reported them must have had a strong impression as to their nature, or they would not have given the name of a disease which was not supposed to exist in that island. These cases presented themselves, and were recorded before the "Bann's" arrival at Ascension; there is no question as to importation, and as they are in accordance with what has been observed frequently elsewhere, of sporadic cases of yellow fever appearing under favourable circumstances, the evidence they afford seems trustworthy.

A distinction as to causation is frequently attempted to be drawn between sporadic cases and those which occur in greater numbers, but without good ground for the opinion. If one case of yellow fever arise in a given situation, it must be admitted that there was a sufficient cause for the disease; the cause may have been slight, and the subject particularly susceptible, still, the nature of the cause must have been such, that if more intense, and more extensively applied, a greater number would have been attacked. It is characteristic of a pandemic wave, that, during its passage, local causes, which under ordinary circumstances seem to exercise inconsiderable influence over the health of those exposed to them, then display a potency, which, if regarded without due weight being given to the reigning pandemic influence, seems quite unaccountable.

Proceeding now to the period 1824-29, it will be seen by Table III., that the mortality in Ceylon was five times greater than in 1823, and the rates at Malta, Gibraltar, and in the West Indies, were much increased, while those in the Ionian Islands and Bermuda were maintained. As in 1818 there was a slight increase in Jamaica, and a low rate at New Orleans, but the ratios of 1823, in Nova Scotia and Canada, were much reduced. There was an augmented ratio at the Cape; that at Mauritius was unaltered, but on the Coast of Africa

Burnett, "Report on the Fever in His Majesty's ship 'Bann," pp. 7-10. +"Climate and Diseases of the African Station," p. 193.