SOILS FERTILIZERS. 219 



soils but also to test the accuracy of the methods employed and the lufluence 

 of other experimental conditions, particularly depth of the soil layer. 



Air dried samples of adobe soil " were sifted through the 0.5 mm. sieve and 

 distributed in duplicate in 1-gm., 3-gm., and 6-gm. portions in weighing bottles 

 which were about Its in. in diameter and lil in. in height. They were 

 arranged on a glass plate which was placed on a porcelain acid dish partly 

 filled with distilled water and set in turn on a ground-glass plate, over the 

 whole of which was fitted a bell jar of the short, squat form, to the inside 

 upper portion of which was pasted a considerable quantity of thoroughly 

 moistened filter paper. The ground rim of the bell jar was well covered with 

 vaseline so as to permit of no escape of the moisture from the bell jar and 

 thus insure a thoroughly saturated atmosphere. The apparatus was then 

 placed on a shelf in a small room as far as possible from currents of air and 

 where the temperature could be raised at will by lighting a Bunsen burner. 

 A centigrade thermometer was placed next to the bell jar so that the tempera- 

 ture could be taken accurately just previous to removing the bottles from the 

 jar for weighing. From the amounts of soil placed in the weighing bottles 

 we had a depth in the case of the 1-gm. samples of 1.5 mm., in that of the 

 3-gm. samples of 4 mm., and in that of the 6-gm. samples of 8.5 mm. The 

 covers of the weighing bottles were protected from moisture and dust under 

 a separate bell jar and were so arranged on a glass plate as to allow of their 

 being placed in their respective bottles in two or three seconds from the time 

 when the jar in which the samples were exposed was raised fi-om the ground 

 glass plate, thus preventing any loss of moisture." 



It was found that by using very thin layers of soil this method gave accu- 

 rate results and it was, therefore, employed in a study of effects of variation in 

 temperature. The more important conclusions reached from this study were 

 that a layer as nearly as possible 1 mm. in depth gave the best results. 

 " Hilgard's method for the detern'inations is recommeuded to be used as a 

 standard in all soil work on hydroscopicity. A rise in temperature is accom- 

 panied by a greater absorption of hygroscopic moisture, a fall in temperature 

 by a decreased absorption. These do not take place according to any definite 

 law. For the arid regions in particular, the hygroscopic moisture in soils has 

 a certain definite practical importance. Similar results on the effect of tem- 

 perature on the absorption of hygroscopic moisture by soils are obtained in 

 the incubator as in the room. The total absorptions in the incubator, how- 

 ever, as well as the variations with temperature are much smaller than in 

 the room." 



Contributions to the study of the mechanical analysis of soils and of the 

 determination of outer soil surface by heat of wetting and hygroscopicity, 

 K, Pfeiffer {Beitrdge zur Frage der mechanischen Bodenanalyse und der 

 Bestimmung der Bodenoherfldche mittels Benetsungsiodrme und Hygrosko- 

 pizitdt. Diss. Univ. Giessen, 1911, pp. 55, fig. 1; Landio. Jahrh., ^1 (1911), 

 No. 1, pp. 1-55). — The author reviews in detail previous investigations on the 

 physical properties of soils as determined by mechanical analysis by sieve and 

 sedimentation, heat of wetting, and hygroscopicity, and reports the results of 

 his own experiments in which he sought to compare the relative values of 

 these three methods in so far as they give an insight into the physical properties 

 of soils. 



From the results of his studies the author concludes that the method of de- 

 termining the surface area by heat of wetting does not commend itself for use 

 not only on account of the complexity of the apparatus required and the de- 

 crease of surface area of the soil upon drying, but because the heat of wetting 



