HQKTICULTURE. 639 



tlirough a glass tube against an air current and are collected at the bottom 

 of the tube in a cup, the larger and heavier seeds thus appearing at the bottom 

 of the collecting vessel. 



Appendix. — x\. brief histoi'y of the tobacco division and its principal experi- 

 mental and breeding work accompanies a discussion of the principal Canadian 

 tobacco growing centers. 



Tobacco crop, 1911; by types and districts, J. P. Killebrew (U. S. Dept. 

 Agr., Bur. Statis. Circ. 27, pp. 8). — A table states the acreage, yield per acre, 

 and total production in 1909, 1910, and 1911 of (1) cigar, and (2) chewing, 

 smoking, snuff and export tyi>es of tobacco, the data being given for each of 

 the principal tobacco-growing States and districts. Another table presents "by 

 States and geographical groups of States the acreage, production, and value 

 of tobacco as reported by the Census for 1909 and 1S99. The crop conditions 

 of 1911 and the quality of that year's crop in each of the principal tobacco 

 States and districts are briefly outlined. 



Notes on some wheat-breeding experiments at the botanical experiment 

 station, Pretoria, J. A. T. Walters (Agr. Jour. Union So. Africa, 2 (1911). 

 No. 6, pp. 765-777, figs. 10). — The author gives brief notes on 6 years' work 

 conducted for the purpose of (1) securing pure strains of local wheats, and 

 (2) improving the existing breeds by the selection of the best strains. 



Wheat manuring- experiments, 1911 {Queensland Agr. Jour., 27 {1911), 

 No. 6, pp. 289-291). — These pages state the results of tests of nitrolime, super- 

 phosphate, sulphate of potash, stable manure, and a proprietary fertilizer, 

 singly and in various combinations, in growing wheat on sandy soils. 



HORTICULTURE. 



Thirteenth report of the Woburn Experimental Fruit Farm, Duke of Bed- 

 ford and S. U. Pickering {Woburn Expt. Fruit Farm Rpt., 13 {1911), pp. 

 151 + VII+258-284+11, pis. 10, figs. 9). — In continuation of previous investiga- 

 tions additional studies relative to the effect of growing fruit trees in grass 

 land are reported and the results secured from the work as a whole are re- 

 viewed. The previous experiments dealt with dwarf and standard apples. 

 (E. S. R., 15, p. 474). In the recent work, the experiments have also Included 

 pears, plums, cherries, and forest trees. Various pot experiments with trees 

 were likewise conducted to test out numerous suggeslions relative to the cause 

 of the stunting effect of grass on trees. 



Summing up their work for the past 16 years, the authors And that the 

 action of grass on fruit trees is often so deleterious that it arrests all growth 

 and often causes the death of the tree. None of the trees thus stunted have 

 recovered from the effect, except in cases where the roots began to extend be- 

 yond the grassed area. The stunting effect is greatest where the ground around 

 young trees is grassed over immediately after planting. Trees where the 

 ground becomes grassed over gradually during the course of several years ap- 

 parently accommodate themselves to the altering conditions and suffer much 

 less than when the grass is actually sown over their roots. Nevertheless, the 

 grassing over of the ground around well-established trees has sometimes proved 

 very detrimental to the growth of the tree. 



Some varieties of apples were found to be more susceptible to injury than 

 others, but no difference was found between the standards on the free stock 

 and dwarfs on Paradise. Pears, plums, and cherries were likewise affected, 

 though with these trees the standards suffered less than the dwarfs. 



The conditions of these experiments have been such as to emphasize the in- 

 jurious effect of grass. The authors are of the opinion that in some soils, 

 38038°— No. 7—12 4 



