The path, I say, if you follow it downhill, will save you, but if, when you find you are in the wrong valley, you attempt to recover your track by going up the lateral ridge, you always run a grave risk. It is by experiments of that sort that men die from exhaustion. It is true that one is not usually tempted to this extra effort. It is much easier to go on the way one is going, and to follow the path down, though one knows it is a wrong one, but there are occasions, especially late in the day, when one has all but conquered the main crest of the range, after perhaps one failure, and when one knows that one is lost, when the idea of one vigorous effort to get over while it is yet daylight is tempting. It is a fatal temptation.

When you have made up your mind that you are lost, or even when the map has told you so, pay no attention to anything else about you or within you, such as the guess that such-and-such a rock in front of one may hide such-and-such a village, or the hope that your strength will hold out for 12 or 15 hours without food, but at once behave like a person in grave danger, that is, calculate your chances of retreat, and think of that only, for I repeat, it is more easy to die from exhaustion than in any other way in these hills, and nearly all the people that perish in mountains perish from that cause.

When you have made up your mind that it is your business to find men again, and that you do not know how far men may be, first note your bread and wine and the rest, if any provision is left; next determine to reserve it until nightfall: eat it then, do not blunder on through the darkness (it is astonishing what very little distance one makes after sunset, and every half-hour of twilight makes it more difficult to camp)—sleep, and take the first half of the next day without food; you are reserving your very last rations until the noon of that day. For one can do a considerable distance without food if the effort is made in the early morning.

Never bathe under such conditions of fatigue, and towards the end, when you are exhausted, drink as sparingly as possible.

It is perhaps useless to give any hints about what a man should do when he is lost, because men get lost in mountains by hoping against hope and pushing on when common sense tells them to return. But I write down these hints for what they are worth. After my first bad lesson in the matter, I found them fairly useful. Remember, by the way, if you are lost and if there is no path apparent, that a cabane even in ruins somewhere in the landscape means a track visible or invisible, and that any rude crossing of the stream with stepping-stones or a log means the same thing. But you must not imagine that the presence or traces of animals will prove a guide, for even mules wander wild for miles on these mountains in places where a man can only go with difficulty and along random tracks leading nowhere.

VI
THE SEPARATE DISTRICTS OF THE PYRENEES

For the purposes of travel upon foot, the range of the Pyrenees falls into certain divisions, which are not very clearly marked, but which arrange themselves in a rough manner under the experience of travel. As I come to deal with each of these, it will be seen that there is not one which does not overlap its neighbour, and it will be impossible to describe any mountain district without admitting this overlapping to some extent, because any valley connected by certain local ties with the valleys to the east and west is also, as a rule, connected with the valleys to the north or south of it. Still, the districts I speak of are fairly distinct, and consist in (1) the Basque valleys, (2) the Vals d’Aspe and d’Ossau, with the valleys of the Aragon and Gallego to their south, which I will call “the Four Valleys,” (3) the Sobrarbe, (4) the three valleys attaching to Tarbes, to which I also attach the Luchon valley, (5) the Catalan valleys and Andorra—in which I include the Val d’Aran, (6) the Cerdagne (omitting the Tet and Ariège valleys), (7) the Ariège and Tet valleys, (8) the Canigou.

These I will take in their order, and I will begin with—