At the eastern side the conditions are a little different, and show us that the mere presence of the sea is not sufficient to produce a "maritime" climate. The prevailing winds in eastern North America are off the shore; they cannot therefore carry oceanic influences landwards. To the north the winds tend to have a northerly component, and cold currents of water also stream out of the Arctic and chill eastern North America. The result is that we find that Labrador, though lying in the latitude of Great Britain, has a very severe climate. Further south the conditions are of the "continental" character even on the seaboard, the winters being very cold and the summers hot. Rainfall is equally distributed throughout the year, but on passing inland it diminishes in amount and tends to be limited to the warm season. The diminution would be much more obvious than it actually is were it not that the existence of the large Gulf of Mexico, and also the size of the North American continent, give rise in the south to a monsoon effect, which greatly increases the rainfall of the south-eastern corner of the States. Further to the west, in the lee of the great barrier of the Rocky Mountains, the rainfall is slight. Incidentally, we may notice that the eastern seaboard of the great Eurasian continent also has a more extreme climate than the western, offering in this respect an analogy to the conditions which prevail on the eastern and western halves of temperate North America. The cause in both cases is the same--the direction of the prevailing winds. We cannot close this chapter without some reference to weather, a subject of more geographical importance than is generally realised. In speaking of climate we have used figures which were invariably _means_, _i. e._ have been obtained by averaging a great number of observations. But where a great number of mean figures are used in a discussion, it is always found that the different averages are based upon varying numbers of observations, and are therefore not strictly comparable with one another. There is always a risk that such figures may mask facts of real geographical importance. No doubt some of the difficulties will disappear with the progress of meteorological science, which will enable the geographer only to select figures which are strictly comparable. Meantime, however, observations for long periods are rare, and the meteorologist must be content to take the figures which are available. For this reason as well as for others, it is advisable to add to the somewhat abstract study of means, that is, of climate, some note upon the actual conditions, that is, upon weather.