440 



SCIENCE 



[N. S. Vol. LI. No. 1322 



Expressing the interception, loss in terms of 

 depth on the horizontal projected area shadowed 

 by the vegetation, the loss per shower of a given 

 amount is very nearly the same for various broad- 

 leaved trees during the summer season. . . . The 

 interception loss from needle-leaved trees, such as 

 pines and hemlocks, is greaiter both as regards 

 interception storage and evaporation during rain 

 than from broad-leaved trees. 



Data are insufficient for a final determiuation of 

 the relative losses from trees in winter and in 

 summer. Apparently the winter and summer 

 losses for a given monthly precipitation for needle- 

 leaved trees the winter interception loss appears to 

 be about 50 per cent, as great when the trees are 

 defoliated as during the growing season. The 

 average interception loss from 11 trees . . . dur- 

 ing the summer of 1918 was 40 per cent, of the 

 preeipitation. 



ATMOSPHERIC MOISTURE IN THE UNITED STATES 



Three years ago, Mr. P. C. Day, chief of 

 tiie climatological division of the Weather 

 Bureau, published a mouogi-aph on " Relative 

 humidities and vapor pressures over the 

 United States, including a discussion of data 

 from recording hair hygrometers,"* and to 

 which recently Mr. W. J. Bennett, of Tampa, 

 Florida, has added an interesting discussion 

 of tables prepared along similar lines for 

 Tampa.^ The diurnal changes in relative 

 humidity (which is the water vapor present in 

 the air divided by the maximum which would 

 be possible at the temperature) are practically 

 the opiKisite of the temperature changes, there 

 being a change generally of 3 to 4 per cent, 

 for each change of 1° C. in temperatm-e. 



Vapor pressure (the pressure exerted by the 

 water vapor locally in the air) is a direct 

 index of the absolute humidity (water vapor 

 per unit volume of space). In summer in dry 

 climates, such as at Boise, Idaho, the vapor 

 pressure rises during the few hours imme- 

 diately after sunrise as the moisture from the 

 surface (e. g., dew) is evaporated. Then after 



* Monthly Weather Review, Suppl. No. 6, 1917, 

 61 pp. (mostly tables), 34 charts. Cf. review in 

 Geogr. Rev., February, 1918, Vol. 5, pp. 155-156. 



5 Monthly Weather Review, July and October, 

 1919, Vol. 47, pp. 466-468, 710, 2 figs. 



about 10 A.M. the vapor pressure decreases as 

 conveetional currents reach higher and higheir 

 and mix the lower air with the drier air above 

 until the principal minimum is reached at 

 about 6 P.M. After this, evaporation, even 

 though small is able to raise the vapor pres- 

 sure in the absence of convection. In a 

 moderately humid climate, such as that of 

 Coliunbus, Ohio, the maximum in summer 

 comes at about 10, as in the drier region, but 

 the minimum is not reached until sunrise, 

 when cooling has condensed a maximum of 

 the water vapor. In a marine climate, using 

 San Francisco as typical, the vapor pressure 

 depends almost entirely on the temperature, 

 and so the maximum comes at about 2 p.m. 

 and the minimum around sunrise. 



In the annual period the relative humidity 

 is usually highest with the lowest tempera- 

 ture; but the vapor pressure varies directly 

 with the annual temperature changes. The 

 vapor pressure is 2 to 4 times as great in 

 summer as in winter in most of the United 

 States. The distribution of relative humidity 

 depends, (1) on the temperature of the air, 

 (2) on the proximity of the main source of 

 moisture, (3) on the prevailing wind direction, 

 and (4) on the topography to windward. East 

 of the Rockies, April is generally the month 

 of lowest relative humidity; while west, the 

 mid-summer months are driest. In most of 

 the United States, the highest relative humid- 

 ity comes in the colder months, except in the 

 southeast where it may occur in late sunmier 

 or early fall. The lowest relative humidities 

 occur in the far southwest, and in the lee of 

 high mountains elsewhere, while the highest, 

 occur near the oceans, similarly, on the lee 

 shores of the Great Lakes, and on the wind- 

 ward sides of mountains. On Pikes Peak and 

 Mount Washington the humidities are gen- 

 erally high and show little variation. In the 

 western half of the country the record minima 

 range from 2 to 10 per cent., while in the 

 eastern half, the lowest are 10 to 20 per cent. 



Since absolute humidity is controlled by 

 temperature more than by any other factor 

 for most of the country, the lowest vapor 

 pressure comes in winter, and is experienced 



