WATEK SOILS. 639 



defeuU'(l tabular record of daily ol nervations on atmospheric pressure, temperature, 

 preci2)itati()n, humidity, cloudiness, direction and velocity of the wind, and casual 

 phenomena. 



Rainfall at the Manurial and Variety Experiment Stations, Barbados, 

 from Dacember, 1900, to April, 1902 [lljd. Agr. Worl; Imp. Dcpt. gr. West 

 Indies, 1900-1902, pp. 4, 121, 122). — The monthly rainfall at 14 different places in 

 Barbados is reported. 



Composition of Barbados rainfall {Rpt. Agr. Worl:, Imp. Dc]>t. Agr. West Indies, 

 1900-1902, p. 2). — The total amount of rainfall and its content of chlorin and of 

 nitrogen in different forms are reported for eacli month from December, 1900, to May, 

 1902, inclusive. 



WATER SOILS. 



A laboratory study of the percolation of water through soil, A. \. Btiiben- 

 KAUfU [CaHfornia SUi. Rpt. 1S99-1901, pt. '..', pjt. 153-172, Jigs. 7). — The percolation 

 experiments reported "were cijnducted in the usual way, i. e., in glass cylinders tilled 

 with the soils (air-dried) under investigation, the rate of flow being easily obtained by 

 measuring the progressive wetting of the soil." In these experiments tests were 

 made of the influences on jjercolation of the diameter of the tubes (5, 2g, |, and ,;^ in. ) ; 

 wetting the soil and drying it at 100° C. and allowing to dry normally; repeated wet- 

 ting and drying; the use of a vent-tube or a cloth bottom to the cylinders; different 

 methods of filling the cylinders; and tamping or keeping the surface loose. 



The results show that the rate of flow is not materially affected by the diameter of 

 the cylinders, but it is difficult to secure uniform results with cylinders of small 

 diameters. Drying at 100° C, repeated wetting and drying either at 100° V. or nor- 

 mally, and tamping the surface retarded percolation. As a result of these observa- 

 tions the author recommends cylinders of l\ in. diameter and 40 in. long. The 

 following arrangement was found satisfatory: "The bottom of the cylinder waB 

 sunk to the depth of one-quarter inch into a thick (2 in.) block of wood, by first 

 boring out a hole of the same diameter as the outside of the cylinder. In the Itottom 

 of this hole a smaller one corresponding to the inside diameter of the cylinder 

 was bored to a depth of 1 in. This smaller hole thus formed a cavity beneath the 

 cylinder, from which an outlet tube was provided. The cavity in the block was well 

 paraffined and waxed to make it water-tight, and filled with coarse fragmentsof granite, 

 over which was pasted a jierforate*! disk of filter paper. The cylinder [provided with 

 a cloth bottom] was filled to the proper mark and secured on top so as to allow it to 

 be inverted. The cloth bottom was remove<l, and the block of wood put in jdace — 

 also inverted. Then by simply turning the whole arrangement upright, the column 

 of soil rested on the layer of granite fragments, unexposed to light and with a 

 minimum exposure to air. 



"The outlet tube led into a weighed fiask, provided with a vent-tube drawn to a 

 fine opening to avoid evaporation while allowing the air to escape. The water was 

 maintained automatically at a depth of one-half inch on top of the soil by means of a 

 Mariotte arrangement; thus evaporation was avoided, and entrance of air at the top 

 end of the < ylinder permitted as in receiving flasks. . . . 



"The old method of filling in the soil piecemeal was discarded as likely to lead to 

 nonuniform conditions in the cylinders. Instead, the entire rj^iantity of soil was 

 weighed out and placed on a sheet of stiff paper a trifle longer than the cylinder. 

 There it was thoroughly mixed and spread out in a long, narrow strip. The paper 

 was rolled up around this strip of soil, forming a cylinder small enough to fit into 

 the glass tube. By pushing the paper roll of soil al^ far as possible into the cylinder 

 lying liorizontally, and then quickly turning the whole into a vertical position, the 

 soil was dropped in the cylinder in one mass. The paper was then carefully and 

 slowly drawn out, the cylinder being tapped gently all the while. In this way the 



