January 25, 1894 J 



NA rURE 



303 



areas covered with spruce will be moister, as well as cooler, 

 than those under woods of less densely foliaged species of trees. 

 In Bavaria it was found that in summer, in consequence of the 

 density of the foliage in beech forests during the most active 

 period of growth, the difference even amounted to 13-6 per cent. 

 of saturation over the relative humidity in the open. 



4. As regards t/ie Pftripiiaiion of Aqueous Vapour. — It has 

 been shown above, not only that the atmosphere within the 

 forest is cooler than in the open, but also that the temperature 

 of the trees themselves is lower, especially in sununer, than the 

 air surrounding them ; hence, when a current of air is wafted 

 from the open into the forest, and comes in contact with the 

 cooler trees, its temperature is reduced, and it is brought nearer 

 to the point of saturation, i.^. its relative humidity increases. 

 But if this air was already in the open at, or near, the point of 

 saturation, then the effect of the cooling process is that a certain 

 amount of surplus moisture beyond the aqueous vapour that 

 can be held by the air up to a point of saturation at its reduced 

 temperature must be released and precipitated in the form of 

 dew. Woodlands, therefore, act as condensers of atmospheric 

 moisture, and decrease the absolute humidity of the air whilst 

 increasing its relative humidity ; and in addition to this, they 

 increase the humidity of the air by transpiration from the leaves, 

 whilst the sap is being rendered available for structural pur- 

 poses, and the work of assimilation is proceeding. 



Endeavours have been made to establish, by means of careful 

 observations, the effect of forests in regard to the precipitation 

 of aqueous vapour in the form of dew or rain, but the results 

 are often of so conflicting a nature that, up till the present, safe 

 deductions caiiDOt be drawn. In order to compare observa- 

 tions made in the forests with those made at the usual meteoro- 

 logical stations m the open, a correction would in each case be 

 necessary to reduce the localities to the same sea-level, as air 

 cools in rising and increases in relative humidity, i.e. it 

 approaches the point at which it must precipitate some of the 

 aqueous vapour held by it. Hence rainfall generally increases 

 with the height of a locality above the sea-level, although no 

 direct proportional increase can be proved. It fluctuates with 

 the geographical position and the varying physical conditions 

 of each point of observation, whilst variations in the direction 

 of the moist winds of the locality also militate against the 

 collection of reliable data for comparison with readings made 

 in other localities. 



The mean of the readings at 192 points of observation in 

 Germany, corrected as carefully as possible with reference to 

 these causes of difference, do not seem capable of giving any 

 more exact inference than the general statement, that at high 

 altitudes large extents of forest may considerably increase the 

 local rainfall. As regards the quantity of rainfall and snow- 

 fall which is intercepted in forests by the leaves, branches, and 

 stems of the trees, the observations made in Switzerland, 

 Prussia, and Bavaria show that nearly one-fourth of all the 

 precipitations of aqueous vapour is intercepted by the forest 

 trees, and is given off again by evaporation, or is gradually 

 conducted down the stems to the soil. In lofty forest-clad 

 regions the mechanical action of the rains on the surface-soil 

 is thus very much modified. 



By means of their lower temperature, their greater relative 

 humidity, and the mechanical obstruction they offer to the 

 movements of currents of air, extensive forests act decidedly as 

 condensers of the aqueous vapour contained in the atmosphere, 

 and their influence in this respect is more marked at high 

 altitudes and in mountainous districts than on plains or near 

 the sea-coast, where other physical factors come into competi- 

 tion with and modify it. Further data are still requisite to 

 enable us to determine with anything like certainty that forests 

 directly cause increase of precipitations irrespective of such 

 local considerations as the ruling direction of winds and pecu- 

 liarities of situation ; the generally accepted dictum is, however, 

 that in the vicinity of extensive forests the rainfall is greater than 

 at other localities under otherwise similar physical conditions. 



In portions of the Russian Steppes, planted up nearly 50 

 years ago, the inhabitants assert that the summer rainfall has 

 considerably increased, and that the danger to crops from 

 drought is not so great as formerly, whilst the villages are also 

 protected by the forest from the violence of the winter storms. 

 In summarising and criticising this point Prof. Endres of 

 Karlsruhe remarks as follows^ : — 



'Conrad, Elster, Lexis, and Loening's 

 wissenschafcen," 1S92, vol. iii. page 6oS. 



NO. 1265. VOL. 49] 



'■ Handworterbuch der Staats- 



" The data furnished from tropical countries must be accepted 

 with the greatest caution, and in any case they afford no con- 

 clusive deductions for European circumstances. Blandford 

 reports from India (Meteorological Journal, 1888) that in an 

 area ot 61,000 square miles, which was formerly denuded of 

 woodlands, but has been planted up again from 1875, ^he 

 rainfall has increased 12 percent, since then. But H. Gannet 

 (ll'ealher, yo\. v.) arrives at exactly the opposite conclusions 

 for America, as his observations in the prairie region and in 

 Ohio go to prove that the re-wooding of a tract exerts no per- 

 ceptible difference on the amount of the aqueous precipitations. 

 Lendenfeld also tries to prove that the clearance of woodlands 

 in Australia has resulted in a better climate and an increase in 

 rainfall, as the soil under eucalyptus remains hard as stone 

 and inabsorptive, whilst it is rendered lighter and more porous 

 by grass. ( Peter/nanii' s Geog. Mittkeilungen, volxxxiv. )." 



5. As regards Evaporation of Soil-RIoistiire. — The low tem- 

 perature and the high relative humidity of the atmosphere in 

 forests are unfavourable to rapid evaporation, which is still 

 further reduced by the protection afforded to the soil against direct 

 insolation and the action of winds. From observations extending 

 over 10 years (1876-85) in various parts of Germany and Austria, 

 the following relation is shown between evaporation in the 

 forests and in the open in the vicinity of the forests : the differ- 

 ences would probably be greater if comparisons had been made 

 with places in the open that were far removed from the modify- 

 ing influence of the woodlands : — 



The practical import- 

 ance of this will be 

 seen, when it is re- 

 collected that the 

 mineral food in the 

 soil can be taken 

 up by the rootlets 

 only in the form of 

 soluble salts. 



It was also found that the amount of evaporation depended 

 on the class of forest, thus : — 



Percentage of Water. 



Species of Woodland. 



Evaporated in Remaining i 

 the Forest. the Soil. 



Beech ... ... ... ... 40 ... 60 



Spruce 45 ... 55 



Scots pine... ... ... ... 42 ... 58 



Clearing for reproduction ... 90 ... 10 



In these statistics no account has been taken of the quantity 

 of water given to the air by transpiration through the leaves ; 

 but this is not of essential importance, as such supplies of 

 moisture are drawn by trees, except during the earliest stages 

 of growth, from the deeper layers of soil and subsoil not im- 

 mediately and directly affected by the aqueous precipitations 

 on the surface. This may be less true of spruce than of other 

 trees. 



The action of forests, therefore, is to retain in the soil a 

 large proportion of the rainfall or of the moisture arising from 

 the melting of snows, which, by percolation to the lower layers 

 and the subsoil, tends to feed the streams perennially, and to 

 maintain a constant supply of moisture, without which trees 

 could not derive their requisite food-supplies from the soil. 



The nature of the soil-covering below the forest trees exerts 

 alsoconsiderable influence on the amount of moisture evaporated. 

 From experiments conducted during five years in Bavaria it was 

 found that a good layer of fallen leaves, and of hunm: or vege- 

 table mould formed by their decay, is a powerful factor ; it 

 diminishes the evaporation by more than half, or reduces it to 

 less than one quarter of that in the open, and thus adds 

 very considerably to the surplus amount of moisture retained in 

 the soil. 



6. As regards the Feeding of Streams and the Protection of the 



