Of 198 genera peculiar to temperate South America, the large majority belong exclusively to Central Chili, and these include several tribes whose affinity to the forms of other regions is only remote. Two of these tribes—the Vivianeæ and Francoaceæ—have even been regarded by many botanists as distinct natural orders; and many of the most common and conspicuous species will strike a botanist familiar with the vegetation of other regions of the earth as very distinct from all that he has known elsewhere. With only a few exceptions, these endemic types appear to have originated in the Andean range, whence some modified forms have descended to the lower country; several of these, as was inevitable, have been found on the eastern flanks of the great range, and it is probable that further exploration will add to the number; but it is remarkable that as yet so large a proportion should be confined to Chilian territory.
Grisebach has fixed the limits of that which he has called the transition zone at the Tropic of Capricorn to the north, and the thirty-fourth parallel of latitude to the south; but these in no way correspond to the natural boundaries. As I have already pointed out, the flora of the desert zone, extending from about the twentieth nearly to the thirtieth parallel of south latitude, shows a general uniformity in its meagre constituents. It is about the latitude of Coquimbo, or only a little north of it, that the characteristic types of the Chilian flora begin to present themselves, and these extend southward at least as far as latitude 36° south, and even somewhat farther, if I may judge from the imperfect indications of locality too often afforded with herbarium specimens.[19]
DISTRIBUTION OF RAINFALL.
In discussing the causes that have operated on the development of the Chilian flora, the same eminent writer has been misled by incomplete and erroneous information as to the climate of the region in question, and more especially as to the distribution of rainfall, which is no doubt the most important factor. It is true that the peculiar character of the Chilian climate makes it very difficult to express by averages the facts that mainly influence organic life. Between the northern desert region, where rain in a measurable quantity is an exceptional phenomenon, and the southern forest region, extending from the Straits of Magellan to the province of Valdivia, Central Chili has in ordinary years a long, dry, rainless summer, followed by rather scanty rainfall at intervals from the late autumn to spring. About once in four or five years an exceptional season recurs, when rain falls in summer as well as winter, and in which the total fall may be double the usual amount, and at longer intervals, usually after a severe earthquake, storms causing formidable inundations occur, when in a few days the rainfall may exceed the ordinary amount for an entire year. When several such storms are repeated in the same year, we may have a total rainfall of three or four times the ordinary average.[20] Such seasons appear to recur six or seven times in each century, and it is clear that, according as the meteorologist happens to include or exclude such a season in his data, the figures expressing the average must vary very largely. Inasmuch as plant life is regulated by the ordinary conditions of temperature and moisture, we are less liable to error in taking the results which exclude exceptional seasons.
In discussing, therefore, the conditions of vegetation in Central Chili, it seems safe to conclude that the averages given in the following table, extracted from the careful work of Julius Hann, “Lehrbuch der Klimatologie,” are above rather than below the ordinary limit. I find, indeed, that while the average rainfall at Santiago during the twelve years from 1849 to 1860 was 419 millimetres, or nearly 16½ English inches, the average for the six years from 1866 to 1871 was 299 millimetres, or less than 12 inches. It is evident that the indigenous vegetation must be adapted to thrive upon the smaller amount of moisture expressed by the latter figures.
CLIMATE OF CENTRAL CHILI.
The following table, compiled from Hann’s work, gives the most reliable results now available, and shows the mean temperature of the year, of the hottest and coldest months, the extremes of annual temperature, and the rainfall for the chief places in Chili, with a few blanks where information is not available. The maxima and minima do not express the absolute extremes attained during the entire period for which observations are available, but the means of the annual maxima and minima. The temperatures are given in degrees of Fahrenheit’s thermometer.
| Places. | South latitude. | Mean temperature of the year. | Mean temperature of January. | Mean temperature of July. | Maximum temperature. | Minimum temperature. | Rainfall in inches. |
|---|---|---|---|---|---|---|---|
| La Serena (Coquimbo) | 29° 56′ | 59·2° | 65·1° | 53·1° | 1·6 | ||
| Valparaiso | 33° 1′ | 57·6° | 63·1° | 52·5° | 77·9° | 45·0° | 13·5 |
| Santiago (1740 feet above sea-level) | 33° 27′ | 55·6° | 66·2° | 45·0° | 87·6° | 30·4° | 14·5 |
| Talca (334 feet above the sea) | 35° 36′ | 56·5° | 70·2° | 45·0° | 19·7 | ||
| Valdivia | 39° 49′ | 52·9° | 61·5° | 45·0° | 84·0° | 29·5° | 115·0 |
| Ancud, Island of Chiloe (164 feet above the sea) | 41° 46′ | 50·7° | 56·5° | 45·9° | 134·0 | ||
| Punta Arenas[21] (Straits of Magellan) | 53° 10′ | 43·2° | 51·3° | 34·9° | 76·3° | 28·4° | 22·5 |
This table brings out very clearly the influence of the cold southern currents of the ocean and air in reducing the summer heat of the western side of South America; for, while the winter temperatures are not very different from those of places similarly situated on the west side of Europe and North Africa, those of summer are lower by 8° or 10° Fahr., and the mean of the year is lower by 6° or 7° than that of places in the same latitude on the east side of South America. It is also apparent that much of what has been stated in works of authority as to the climate of this region is altogether incorrect. In his great work on the “Vegetation of the Earth,” Grisebach gives the mean temperature of Santiago as 67·5°, or nearly 12° higher than the mean result of ten years’ observation, and the rainfall as over 40 inches, or nearly three times the average—more, indeed, than three times the average—of ordinary seasons.