WATER SOILS. 297 



" Costof j^umping wafer. — In niisint; the water from the hike for these experiments 

 a No. 4 centrifugal jjump was used, driven by a common portalde farm engine. The 

 water was drawn through a 6-inch suction pipe 110 ft. long, raised to a height of 

 26 ft. and conveyed by gravitj- to the reservoir through a 6-iuch galvanized iron 

 conductor pipe. 



"iJuring 3 days the coal used in pumping was weighed and the amount burned on 

 the 8th, 10th. and 11th of September was 1,86.5.1 lbs. of Indiana block, pumping with 

 it 74,909 cu. ft. of water, or at the rate 80,320 cu. ft. per ton of soft coal, eiiual to 

 22^ acre-inches. Had we been provided with a larger discharge pipe and facilities for 

 using water more rapidly than was possible at the time, the same amount of fuel 

 would have been able to raise much more water than it did under the unfavorable 

 conditions of the experiment." 



The rate of percolation from long columns of soil, F. H. King 



{Wisconsin Sta. Rpt. 1894, pp. 285-288, Jig. 1). — Iii contiiiuatiou of 

 iuvestigations noted in a previous report,' tubes 8 ft. long- and 5 in. in 

 diameter ^yere filled with water-free sand of different degrees of fine- 

 ness, viz., that which passed screens with 20, 40, 60, 80, and 100 meshes 

 to the inch, respectively. Each cylinder was filled with water from 

 below until it overflowed at the top. The percolation from these cylin- 

 ders during the period from January 30 to October 24 is recorded. 



"The most remarkable feature of this experiment is the very long period during 

 which the percolation continued, and not much less surprising is the irregularity 

 with which it occurred. All of the cylinders discharged some water as late as July 

 13, after the lapse of 164 days, while one cylinder discharged 69,7 gm. of water on 

 October 24, after a lapse of nearly 9 months. . . . 



"After the first rapid jiercolation had taken place all samples of soil went on dry- 

 ing at nearly equal rates, each losing not quite 1 per cent of its dry weight, except 

 the coarsest sand, which continued to percolate not only longest but also lost the 

 largest amount of water during the latter interval, as it did during the first. . . . 



"These observations have a very important bearing upon the irrigation of sandy 

 lauds, and show in an emphatic manner that it would be extremely wasteful not only 

 of water but also of fertility to apply water in large quantities at a time, unless 

 these lands are underlaid at a shallow depth with a much more impervious layer." 



Results obtained in the culture of Swedish marsh lands, 

 C. VON Feilitzen {Svenska Mossk. For. Tidskr., l<s!)o, pp. 34.5-353). — 

 The author reviews the work of the Swedish Marshland Association 

 during the past year. Excellent results were obtained in the culture 

 of "high marshes" (sphagnum marshes). These marshes were tile 

 drained and cultivated later in the seasou. During the winter months 

 sand was spread on the drained marsh, followed in the spring by lime, 

 Thomas slag, kainit, and about 46 bu. per acre of inoculated soil. The 

 land may then be sown to peas, vetches, etc., or large crops of clover 

 may be grown. 



Kainit, in the author's experience, appears to have a marked effect 

 against injurious insects. During the past year the grass plats experi- 

 mented on were failures where Thomas slag was applied alone, while 

 heavy yields were obtained on plats receiving kainit in addition to the 



slag. — F. W. WOLL. 



' Wisconsin Sta. Rpt. 1893, p. 175 (E. S. R., 7, p. 565). 



