April 8, 1920] 



NATURE 



159 



increasing- -the general rainfall is probably 

 negligible in the British Isles. 



The greater part of the volume is devoted to 

 a question of national importance — the afforesta- 

 tion of water-catchment areas, with particulars 

 of the extent to which the work has already pro- 

 ceeded. This is timely, for the whole question of 

 the water resources of the country is now under 

 investigation by a Committee appointed by the 

 President of the Board of Trade and the Minister 

 of Health. Prof. Henry shows clearly that the 

 gathering grounds for the reservoirs of water 

 supply by gravitation are well fitted in almost 

 every case for planting with timber-trees up to 

 an elevation of 1000 ft., and he adduces some 

 evidence to show that covering a certain pro- 

 portion of the surface with forest growth, so 

 far from being detrimental to the yield of the 

 areas, is even helpful. Curiously enough, he does 

 not refer to the important influence of high vegeta- 

 tion, and especially of trees, in precipitating 

 moisture from mist, a phenomenon which is shown 

 very strikingly when one is traversing- a road 

 through a wood in a thick mist. The road 

 remains perfectly dry, while the drip of water 

 from the branches on either side gives out the 

 sound of abundance of rain, and the ground be- 

 neath the trees soon becomes saturated. As Dr. 

 i Marloth proved on Table Mountain many years 

 ago, even such inconspicuous growths as a bed of 

 reeds can draw liquid streams from a cloud drift- 

 ing along- mountain slopes which would deposit 

 nothing on a bare stretch of soil or rock. No 

 doubt a certain amount of water is in this way 

 added to a forest-covered catchment area without 

 being recorded in properly exposed rain-gauges. 



On the other hand, if, as certain experiments 

 made in Germany and quoted on p. 3 seem to 

 suggest, the transpiration of forest trees is greater 

 than the evaporation from an exposed water 

 surface, the net result may be to reduce the 

 amount of water reaching a reservoir, and this 

 might be a serious matter in a dry summer. Even 

 if we admit that afforestation does not appreciably 

 increase the available run-off, it seems unlikely 

 that it can seriously diminish it, and the balance 

 of probability is that planting a water-catchment 

 area is beneficial. A wide belt of woodland sur- 

 rounding a reservoir must reduce the wash of 

 surface material into the streams, and so retard 

 the silting- up of the reservoir. But, what is 

 more important, the value of the forests when 

 once established will justify the acquisition of the 

 whole drainage area of their water supply by 

 authorities which could not otherwise justify the 

 expense of such a step ; and it is only on ground 

 NO. 2632, VOL. 105] 



which is the property of the water authorities that 

 it is possible to keep the area free of population 

 or of farm stock, and so secure the water abso- 

 lutely from all sources of pollution. 



This is scarcely the place to criticise the list of 

 catchment areas for water supply in the United 

 Kingdom, which fills 135 pages, and appears to 

 be based on official figures ; but one cannot help 

 regretting that the data quoted are not more 

 homogeneous. Rainfall figures, for instance, are 

 given only in some cases, and even then they are 

 often misleading from the lack of information 

 as to how the average was computed. The 

 responsibility for this is on the local authorities 

 themselves, and we can only thank Prof. Henry 

 for his diligence in bringing so many facts 

 together that the room for improvement in the 

 form of statement stares one in the face. No 

 remedy for these ill-assorted statistics can be 

 found until some central water authority comes 

 into existence which can co-ordinate all the local 

 efforts in accordance with one national and scien- 

 tific system. One slip, however, should be 

 corrected — the allocation of Haweswater to the 

 supply of Penrith on p. 175 and on the map on 

 p. 173. This should be Hayeswater, a small 

 tarn lying between Haweswater and Ullswater. 

 HuGH Robert Mill. 



The Wilds of South America. 



(i) In the Wilds of South America: Six Years of 

 Exploration in Colombia, Venezuela, British 

 Guiana, Peru, Bolivia, Argentina, Paraguay, 

 and Brazil. By Leo E. Miller. Pp. xiv -1-428. 

 (London: T. Fisher Unwin, Ltd., 1919.) Price 

 215. net. 



(2) University of Pennsylvania. The University 

 Museum Anthropological Publications. Vol. ix.. 

 The Central Arawaks. By William Curtis 

 Farabee. Pp. 288+xxxvi plates. (Philadelphia : 

 The University Museum, 1918.) 



(0 IV /T^- MILLER'S well-illustrated book is the 

 iVl attractively written personal narrative 

 of seven exploring expeditions to South America, 

 all but one sent out by the American Museum of 

 Natural History, mainly to collect mammals and 

 birds. 



There are few wild countries left which have 

 not been ransacked with the hope that the 

 amassed specimens may include some species 

 "new to science," but there are still fewer the 

 fauna and flora of which have been correlated 

 intelligently with scientific observation of the pre- 

 vailing environmental conditions. Several Ameri- 

 can museums stand in the front rank of such 



