Havana is the most northern point to which the endemic area ever extended. Toward the south the endemic area spread much farther from the original endemic focus than on the north. While in distance from the original endemic area the disease spread further south than north, it did not in point of latitude. Rio Janeiro and Santos, the furthest points south to which the endemic area spread, are about latitude 23°, much the same as Havana on the north. Para, Manaos, Pernambuco and Bahia gradually came within the endemic area, until finally, in 1849, yellow fever took up its permanent habitat in Rio and Santos. During the year 1850, forty-one hundred and sixty persons died of yellow fever in Rio. These cities, like Havana, were subject to epidemics of yellow fever years before they became endemic centers. Endemicity did not spread regularly over this area from north to south. The disease became endemic in Rio thirty or forty years before it did in Manaos. Other factors proved more potent than location as to latitude. The large emigration to Brazil, which set in about the middle of the nineteenth century, gave a constant and large supply of non-immunes to Rio, while Manaos did not get this supply until railroad construction commenced in the interior, about the beginning of the present century. Strictly speaking, Pernambuco, on the eastern coast of Brazil, about longitude 35° west, is the point farthest east where yellow fever really became endemic, in the sense of remaining continuously in one city for a number of years. But on the west coast of Africa, from St. Paul de Loanda on the south to the Canary Islands on the north, yellow fever has been present all the time at some point or points, ever since its introduction in 1494, though on this coast it has never manifested itself continuously in the same town or city for any considerable number of years. But it is endemic on the coast in the sense that it no longer has to be re-introduced from the outside, but is always present at some point. We have a pretty accurate account of its first introduction into the Canary Islands in 1494 by Spaniards returning from Hispanola. The farthest point west at which it has ever become endemic is its original home, Vera Cruz, Mexico. It has been endemic at both Panama and Guayaquil, on the west coast of America. While these points are on the west coast, and Vera Cruz is on the east coast, still, Vera Cruz is some sixteen degrees of latitude west of either Panama or Guayaquil. The endemic limits of yellow fever at its period of greatest extent would be defined by a line drawn from Havana, commencing on the north, to the Canary Islands, down the west coast of Africa to Loanda, from Loanda west to Rio Janeiro, Brazil, from Rio Janeiro to Guayaquil, Ecuador, from Guayaquil to Panama, from Panama to Vera Cruz, Mexico, and from Vera Cruz back to Havana, a very restricted area compared with other infectious diseases.