ON THE PACIFIC COAST 335 



I have never seen either leaves or flowers of the Yuca del 

 monte ; but, from the description given me of it, I should suppose 

 it a Convolvulacea, allied to the sweet potatoes (Batatas), and the 

 lanceolate leaves point to the genus Aniseia. 



The arborescent vegetation of the desert, although perhaps 

 really more scanty than the herbaceous, is from its nature more 

 conspicuous wherever it exists. There are points from which not 

 a single tree is visible all around the horizon, but they are rare ; 

 generally the view takes in a few widely-scattered trees growing 

 in basin-shaped hollows or towards the base of slopes, where at 

 a certain depth there is permanent moisture throughout the wide 

 interval between the anos de aguas, at which epochs the supply 

 is renewed. Wells dug in such sites reach water (too brackish 

 for drinking) at various depths, the first deposit often at only 

 a few feet from the surface. The moisture derived from the 

 garuas, scanty as it is, no doubt aids in keeping the desert plants 

 alive ; and we have already seen that the air is never so ex- 

 cessively dry as might be supposed, but, on the contrary, some- 

 times approaches complete saturation. The trees of the desert 

 are the Algarrobo (Prosopis horrida}, the Vichaya (Cap par is 

 crotonoides\ the Zapote del perro (Coticodendrum scabriduml!\ 

 and an Apocynea with numerous slender branches, bright green 

 lanceolate acuminate leaves, axillary clusters of small white 

 flowers, and fruits, consisting of small twin drupaceous follicles, 

 which are slender, curved, and coated with a thin white flesh. 

 The Capparis and the Apocynea, although they grow to be trees 

 in favourable situations, as in valleys near the sea, are mere 

 shrubs on the desert ; and the Prosopis and Colicodendron are 

 low trees of very scraggy growth, their branches all bent one 

 way by the prevailing wind, and the trunk itself often semi- 

 prostrate. 



Far away over the desert a tall branched Cactus begins to In- 

 met with; the same species abounds on the desert -coast of 

 Ecuador. Farther still, near the roots of the Cordillera, the 

 vegetation becomes gradually more dense and varied, comprising 

 several other kinds of trees, and amongst them most of those 

 about to be mentioned as deni/ens of the valleys. 



When the traveller across the despoblado comes suddenly on 

 one of the valleys, he passes at once from a desert to a garden, 

 \\ lio^e charms are enhanced by their unexpectedness. Standing 

 on the cliff that overlooks the Chira, about Amotape, he sees at 

 his feet a broad valley filled with perpetual verdure, the great 

 mass of which is composed of the pale green foliage of the 

 Al-arrobo ; but the course of the river that winds through it is 

 marked (even where the river itself is not seen) by lines or 

 groups of tall Coco palms, here and there diversified by the more 

 rigid Date palm, both growing and fruiting in the greatest 



