888 EXPERIMENT STATION RECORD. [Vol.43 



the effect of which could be controlled by the design of the flume so as to 

 obtain a more definite coefllcient. 



The maximum loss of head necessary in the use of the Venturi flume is very 

 small compared to that lost with other measuring devices under similar condi- 

 tions. A weir with the same range of capacity would probably involve at least 

 five or six times as -much loss of head. The necessity for connecting each of the 

 gauge wells to a pair of opposite gauge openings in the channel was verified. 

 The exact location of the upstream gauge openings was found to be immaterial 

 as long as they were placed upstream from, and fairly close to, the beginning of 

 the convergence. 



The determination of the discharge of the Venturi flume, from the readings of 

 the upstream and throat gauges, is slightly more involved than in the case of the 

 Venturi meter in a pipe since, besides the Venturi difference, the depth nmsl 

 also be taken into account. For this reason an automatic recording device such 

 as is commonly used with the Venturi meter, while undoubtedly practical, would 

 probably be more complicated. 



Curves and tabular data are included. 



The stopping of sink holes in canals and the construction of drops in 

 very light soil, C. J. Moouy {Rcclam. Rec. [C7. S.], 11 (1920), No. 8, pp. 374- 

 378, figs. 7). — Construction methods used by the U. S. Reclamation Service on 

 the Flathead irrigation project in Montana for the repair of sink holes in canals 

 and the construction of canal drops ai'e described. 



Water supplies and public health {Health Bui., 35 (1920), No. 8, pp. 32, 

 figs. 23). — This bulletin contains considerable popular information on water 

 supplies and their purification, with particular reference to residential and 

 farm water supplies. The appearance of methods and apparatus for the purifi- 

 cation of small farm water supplies, some of which were described by the U. S. 

 Department of Agriculture several years ago, would indicate that little advance 

 has been made on the subject. 



Land clearing, O. I. Bebgh and A. H. Benton (Minnesota Sta., Rpt. Grand 

 Rapids Substa., 1915-1919, pp. 59-70, figs. 8). — This report gives information on 

 a land clearing project, the purpose of which was to determine the relative 

 merits and cost of (1) clearing land of stumps with a horsepower stump puller, 

 (2) shattering the larger stumps with djaiamite before pulling with a horse- 

 power stump puller, and (3) blasting out the stumps with dynamite. The tract 

 of land selected had about an equal mixture of hard and soft wood stumps. The 

 hardwood stumps had been cut from 3 to 10 years and the softwood from 20 to 

 30 years. The soil on the east half of the tract was silty loam and on the west 

 half sandy loam to pure sand. 



The data indicate very conclusively the advantage of pulling over dynamiting 

 in stump removal. Shattering before pulling had a slight advantage over pull- 

 ing without shattering. There did not seem to be much difference in the 

 amounts of 20 and 40 per cent dynamite used, although the 40 per cent blasted 

 out the stumps more effectively than the 20 per cent. An advanage of shatter- 

 ing stumps before pulling was that much less dirt was turned up mth the roots 

 and the parts of the stump were much moi'e easily handled. The cavities left 

 after blasting were much larger than those left where stumps were pulled. It 

 was observed that many more roots were removed when stumps were pulled 

 than where dynamite was used. Another advantage of pulling was that a large 

 number of small stumps were removed that still remained when dynamiting 

 was completed. 



Land clearing and wood utilization by distillation, W. R. James ([WUr 

 mington, Del.]: Hercules Poivder Co. [1920], pp. [1]+S2, pis. 6).— This is a 

 paper delivered at the Agricultural Extension Directors' Convention at Gulf- 



