838 EXPERIMENT STATION RECORD. 



Other tables report the amount of water used per kilogram of plant substJ^nce 

 produced by the plants. ' 



The total nitrogen content in the roots decreased as the amount of water 

 applied increased, but probably no less nitrogen was absorbed by the large 

 heavily watered beets, as the nitrogen rose to the foliage in this case. The 

 decrease in the total nitrogen content was even more rapid than the increase 

 in the amount of water applied, and the quality of the beets was Improved. 

 This is in harmony with the common observation that during dry seasons the 

 injurious nitrogen content is high. 



The composition of sug-ar beets of different sizes, A. Herke (Osterr. Uiigar. 

 Ztschr. Zuckcrindus. u. Landw., Ifl {1912), No. 1, pp. 8-12). — Each of a number 

 of lots of beets was sorted, on the basis of weight, into 3 groups. Tables show 

 the chemical composition of the beets in each group. 



Soaking' beet seed, G. Kock {Osterr. JJngar. Ztschr. Zuckerindus. u. Landw., 

 41 {1912), No. 1, pp. 13-15). — Soaking beet seed 20 hours in pure water gave 

 a quicker, more even germination, healthier and more vigorous plants, and 

 heavier beets. The results of treating with 0.5 per cent carbolic acid are also 

 briefly reported. 



The germination of hulled timothy seed, B. Kajamus {Fi'Ming's Landw. 

 Ztg., 60 {1911), No. 12, pp. J,31-434) .—Tests of 19 samples grown during the 

 j-ear preceding the test showed an average germinating power of 93 per cent 

 for the seed without hulls, and 97.5 per cent for that with hulls. In case of a 

 poorly stored sample the tests were 61.3 and 64.3 per cent, resi^ectively. This 

 seed was probably immature. In another group the author includes 6 samples 

 in which the seed with and without hulls showed average germination tests 

 of 79.5 and 42.9 per cent, respectively. He concludes that storage affects more 

 quickly the seeds which have lost their hulls, and that they are more subject 

 to mold. Their metabolic processes are hastened and the germ dies more 

 quickly. 



Tobacco: Breeding' cigar filler in Ohio, A. D. Sbu^by and T. Houser {Ohio 

 Sta. Bui. 239, pp. 3(U-.'il9, figs. 30). — This bulletin, continuing previous work 

 (E. S. R., 27, p. 237), states the conditions surrounding tobacco improvement in 

 the Ohio tobacco filler district, and records the results in producing new types 

 by selection and hybridization at the Germantown test farm in cooperation with 

 the Bureau of Plant Industry of this Department. The theory and practice of 

 hybridization and tobacco breeding are discussed in detail and the more im- 

 portant hybrids originated in this work are described. 



The selection of individual seed plants of Zimmer Spanish resulted in no 

 progress toward a better quality or a higher yield, and indicated that improve- 

 ment was possible only by the method of crossing followed by selection. In 

 1906, plants from seed 15 years old and which germinated nearly as well as new 

 seed did not differ in character or yielding capacity from other Zimmer 

 Spanish tobacco grown. In 1910, old seed claimed to have passed through but 

 2 generations since the original supply had been obtained from the supposed 

 originator of the variety, produced less than the checks grown with it, and the 

 tobacco again appeared exactly like the ordinary Zimmer Spanish of to-day. 

 These results are believed to disprove the claim that the variety is running out. 



Results obtained in 4 different years with Connecticut Havana led to the con- 

 clusion that this variety is identical with Zimmer Spanish. These results were 

 also regarded as showing a remarkable fixity of type in which even radical 

 changes of environment do not produce inheritable variation. 



In 1906, Ohio Seedleaf and Pennsylvania Seedleaf in a comparative test 

 yielded 44.6 and 40.6 per cent more respectively than Zimmer Spanish. For the 

 4 years 1907-1910, selections of Pennsylvania Seedleaf, numbered 9 and 12, pro- 



