July 23, 1896J 



NA TURE 



28' 



two different routes : the eastern route, Hami to Sa-chou, 

 and the western route, from Kurla to Lob-nor, and thence east- 

 wards to Sa-chou, along the northern foot of the great border 

 ridge, the Altyn-tagh. Two excursions made, for 150 miles, into 

 the interior of the desert gave an insight into its physical 

 features, flora and fauna. Moreover, before crossing the Gobi, 

 the expedition explored in detail the remarkable LuUchun 

 depression (in the south-east of Turfan), which was discovered by 

 the brothers Grum-Grzimailo, and proved to be some 150 feet 

 below the level of the ocean, although it is surrounded on all sides 

 by high plateaus. Roborovsky established there a meteorological 

 station, at which the barometer was observed for two consecu- 

 tive years, and, accordingly, it may now be taken as certain that 

 the surface of this depression is really from 150 to 300 feet 

 below the sea-level. 



Spending nearly one year in the Nan shan highlands, the ex- 

 pedition has covered them with a whole network of surveys ; so 

 that when these surveys, as well as ObruchefTs researches 

 are taken into account (as they are in a preliminary map 

 appended to the Izvestia of the Russian Geographical Society), 

 we see this region, almost entirely unknown three yeirs 

 ago, better explored now than many parts of Siberia. Where 

 one ridge only was formerly drawn, we find on the new map 

 a series of parallel ridges all running W.N.W. to E.S.E., 

 intersected by high valleys, and attaining by their snow-clad 

 peaks the heights of from 14,000 to 16,000 feet in the chains of 

 Humboldt, Ritter, Da-sue-shan, and Alexander III.'s. The 

 beginnings also have been made of a careful exploration of the 

 Altyn-tagh, which was formerly known through Prjevalsky's and 

 Littledale's journeys along its outer border. 



It is pleasant to add that Roborovsky's health has much im- 

 proved during the return journey, and that, on arriving in 

 Russian Turkestan after a two years" absence, he could report 

 "all well." The account of this journey will add several 

 more important volumes to the scientific literature of Central 

 Asia. 



EVAPORATION} 



T^ 



"HE quantities of water added to the atmosphere daily by 

 evaporation from the oceans and the continents constitute 

 a fundamental consideration in meteorology ; the quantities 

 evaporated from cultivated fields, forests, and other forms of 

 vegetation are equally important in agriculture, but as yet we 

 have confessedly attained to only a very imperfect knowledge of 

 this subject. Meteorologists have generally observed the amount 

 evaporated from a small surface of water exposed either in the 

 open air and sunshine, or else within such a shelter as is used 

 for the open-air thermometer ; lately a disc of moist paper has 

 been substituted for the surface of water, as in the Piche 

 evaporometer. Agriculturists, on the other hand, have made 

 use of the lysimeter, which consists of a deep metallic box buried 

 in the earth and having its open upper side flush with the surface 

 of the ground. This box is filled with soil in which plants may 

 or may not be growing, according to the object of the investi- 

 gator. Record is kept of the amount of water or rain that is 

 added to the lysimeter box from day to day, and also of the 

 amount of water that drains from the bottom of the box. The 

 difference between the two is adopted as the natural evaporation 

 from the soil. The soil in the box may be kept very wet, to 

 imitate a morass, or very dry to imitate a desert ; the fineness 

 of the soil may vary from coarse gravel to the finest silt. 



If we desire the actual amount evaporated into the atmo- 

 sphere, we must do more than record the results of the above 

 forms of apparatus. The evaporating surface of water in the 

 shaded thermometer shelter will indeed give up its moisture in 

 proportion to the temperature of the water and to the velocity 

 and dryness of the wind at its surface ; but these three important 

 factors have values so different out of doors from those within 

 the shelter, that such records can, at the best, only give us a 

 crude idea of the actual evaporation from surfaces in the open 

 air. A great evaporation within the shelter, caused by a strong, 

 hot, dry wind, may be accompanied by but little evaporation 

 from the surrounding country if the latter be a desert of rock 

 and gravel. 



On the other hand, by means of the lysimeter, one may 

 indeed determine directly the evaporation from soil of any 

 character exjujsed to the natural outdoor conditions ; but there 

 1 Prof. Ckv<;land Abbe, in thi: U.S. Monthlv Weather Rc-jievj. 



NO. 1395, VOL. 54I 



then remains the difficult task of determining how much soil 

 of each respective kind really occurs in the surrounding 

 territory. In order, therefore, to determine the actual evapora- 

 tion from land surfaces, one must observe a large number of 

 lysimeters, and make an extensive minute survey of the country. 

 The calculations incident to this latter method are very 

 laborious. 



The ordinary psychrometric observations give the dew-point 

 or quantity of moisture in a small unit volume of air at any 

 moment. If in the course of the day this quantity increases, we 

 are not thereby warranted in concluding that the increase is due 

 to a local evaporation ; it may have been brought from a distance 

 by the wind, or it may even have come down from the clouds 

 as rain. If observations of dew-point are carefully made on all 

 sides of a large field, over which a gentle wind is blowing, and 

 if it should appear that there is a little more moisture in the air 

 on the leeward side than on the windward side, one might con- 

 clude, provisionally, that this increase represented the quantity 

 of moisture thrown by evaporation into the air as it gently 

 moved over the surface of the field. But even this conclusion 

 must be modified indefinitely by the consideration that in 

 blowing across the field the wind does not move horizontally, 

 but in a series of rolls and whirls by which the lower air in 

 which we are observing becomes mixed with upper air, about 

 whose moisture we know little or nothing. 



In the midst of all these uncertainties it seems almost hopeless 

 to attempt anything like an accurate determination of the 

 moisture actually added to the atmosphere by evaporation from 

 any extensive region of land or water ; the question is far more 

 complex than the determination of the evaporation from a 

 reservoir of water, which latter problem is often attacked by the 

 hydraulic engineers. Including the earth and its atmosphere in 

 one comprehensive view, we may certainly say that the total 

 annual evaporation from snow and ice, fresh water and salt 

 water, must average the same as the total annual precipitation. 

 We may even make an annual average for each continent, and 

 say that the evaporation from the land plus the water that flows 

 away in the rivers must equal the rainfall, and as the river dis- 

 charge is frequently well known, we may, by subtraction, infer 

 the evaporation. For the oceanic surface, on the other hand, 

 the evaporation must equal the rainfall plus the river discharge 

 from the continents. 



The latest contribution to our knowledge of evaporation from 

 land surfaces is published by Prof. E. WoUny, of Munich, at 

 page 486, vol. xviii., of his " Forschungen." As, the result of 

 three years' continuous observations of five lysimeters and a 

 neighbouring evaporometer, he conchides : 



(1 ) That the quantity of moisture evaporated from the soil 

 into the atmosphere is considerably smaller than that evaporated 

 from a free surface of water. 



(2) That the evaporation is smallest from naked sand, and 

 largest from naked clay, whereas naked turf and humus or 

 vegetable mould have a medium value. 



(3) That the evaporation is increased to a considerable extent 

 by covering the ground with living plants. 



As the result of a minute analysis of the complex relations 

 between the evaporation and the meteorological elements, on 

 the one hand, and the physical features of the soil, on the other. 

 Dr. Wollny further concludes as follows : 



(4) Evaporation is a process that depends both upon the 

 meteorological conditions and on the quantity of moisture 

 contained by the substratum of soil. 



(5) Among the external circumstances temperature is of the 

 greatest importance, inasmuch as, in general, evaporation 

 increases and diminishes with it ; but this eft'ect is modified 

 according as the remaining factors come into play, and in 

 proportion to the quantity of water supplied by the sub- 

 stratum. 



(6) The influence of higher temperature is diminished, more 

 or less, by higher relative humidity, greater cloudiness, 

 feebler motion of the wind, and a diminished quantity of 

 moisture within the soil, whereas its influence increases under 

 opposite conditions. On the other hand, low temperatures 

 can bring about greater eft'ects than high temperatures if the air 

 is dry, or the cloudiness small, or the wind very strong, or if a 

 greater quantity of water is present within the evaporating 

 substance. 



(7) For the evaporation of a free surface of water, or for 

 earth that is completely saturated with water, the important 

 elements are— first the temperature, next the relative humidity 



