Apeil 3, 19G3.] 



SCIENCE. 



551 



gonia, 1896-1899,' Vol. I., 'Narrative of the 

 Expeditions, Geography of Southern Pata- 

 gonia,' by J. B. Hatcher, Princeton, 1903, 

 4to, xvi + 314 pp., map and numerous plates). 

 The narrative abounds with interesting de- 

 tails of three journeys. The general account 

 of the geography, in chapters headed plains, 

 mountains, rivers, lakes, coast, climate, and 

 Indian tribes, is most readable and instruc- 

 tive, although rather brief on certain topics 

 vcliere additional details v^ould be welcome. 

 The curious relation of the large piedmont 

 lakes, east of the mountains, to the gorges by 

 which they are drained through the main 

 chain of the Andes, is properly characterized 

 as unique; too little consideration seems to 

 be given to glacial erosion in connection with 

 these lakes. The great transverse valleys by 

 which the plains are broken are, for the most 

 part, followed by small or intermittent rivers ; 

 the valleys are shown to have been eroded 

 before the submergence of the region, during 

 the recovery from which the great shingle 

 formation was spread over the plains as a 

 littoral marine deposit. The terraces in the 

 plains are ancient sea cliffs, cut during pauses 

 in emergence, the cliff along the present coast 

 being the last member of the series. Mo- 

 rainic deposits are abundant over the western 

 plains, and extensive lava flows are spread 

 over the central part of the plains; some of 

 the flows are older than the great valleys, 

 some are younger. In one case a river that 

 once followed a valley to the Bay of San 

 Julian was turned southward from its course 

 by a lava flow, so that it now reaches the sea 

 by Eio Chico de Santa Cruz, leaving its 

 former valley dry. The southernmost of the 

 transverse valleys, not yet entirely emerged, 

 forms the Straits of Magellan. The chapter 

 on the Tehuelche tribe gives many examples 

 of the immediate dependence of these savages 

 on their surroundings; they have curiously 

 enough abandoned the use of bows and arrows, 

 remains of which are found in their old camp- 

 ing grounds; since the introduction of horses 

 by the Spaniards, the bolas are the chief 

 weapon of the Indians. 



CAPTURED VALLEYS m THE HIMALAYAS. 



Freshfield, Garwood and Sella made a 

 tour around the highest mountain in the 

 world during the autumn of 1899, and some 

 account of their results have lately appeared. 

 The leader of the party gives a narrative of 

 the trip, with a superb panorama by Sella, in 

 an article on ' The Glaciers of Kangchen- 

 junga' (Geogr. Journ., XIX., 1902, 453^72); 

 and Garwood follows with some " Notes on a 

 Map of ' the Glaciers of Kangchenjunga ' 

 with remarks on some of the physiographic 

 features of the district" {ibid., XX., 1902, 

 13-24). From the latter article we learn that 

 the mountain slopes in the forested belt, up 

 to about 10,000 feet, have ' a marked convex 

 curve produced by the thick growth of vege- 

 tation,' instead of the typical concave basal 

 curve; that the glaciers of the district for- 

 merly extended at least several miles beyond 

 their present ends; that lakes are rare and 

 small ; that the ' entire absence of rock basins 

 from valleys formerly filled by ice is not 

 without bearing on the supposed origin of 

 lakes by glacial erosion in other alpine dis- 

 tricts ' ; and that hanging valleys were ob- 

 served on several occasions in greater or less 

 distinctness. 



Two conspicuous examples of the last- 

 named features are illustrated. They are ex- 

 plained as the high-level valley-heads of a 

 former east-flowing consequent river system, 

 now captured by a deep-lying, south-flowing 

 subsequent stream. The excessive deepen- 

 ing of the subsequent valley beneath its 

 hanging laterals is referred to two causes: 

 (1) A hypothetical elevation of the central 

 mountain mass due to the melting off of 

 former supposedly heavy glaciers during an 

 assumed interglacial period or periods, as 

 a result of which the centrifugal south- 

 flowing subsequent stream would deepen its 

 valley, while the streams flowing ' east and 

 west would be merely tilted sideways, and 

 would tend to widen rather than deepen their 

 valleys'; (2) a postulated protection of the 

 hanging valleys by local glaciers, which 'would 

 linger longer in the high-level hanging val- 

 leys than in the deeper valleys below.' 



Whether the deep subsequent valley was 



