No. 2, June, 1921] GENETICS 157 



1087. Harris, J. Arthur, axd C. S. Scofield. Permanence of differences in the plots 

 of an experimental field. Jour. Agric. Res. 20: 335-356. 1920. — The fact is generally recog- 

 nized that heterogeneity is a source of error in experimental field tests. The purpose of the 

 writers is to show whether the differences among experimental plots are permanent or whether 

 they are transitory and are smoothed out by cultivation. Interannual correlations are used 

 as a means of measuring the relative permanency of the differences. — Published data are 

 reviewed for yields of paddy on 17 plots of ranges "B" and "C" of the wet tract of the 

 Experimental Farm at Hebbel, Mysore. The correlation between yields of the same plots 

 for 1905 and 1906 is for range "B" + 0.834± 0.050 and for range "C" + 0.799 ±0.059. Correla- 

 tions for yields of ragi on 105 plots of the dry-land experiments of the Mysore state 

 for the years 1905 and 1906, 1905 and 1907, and 1906 and 1907, are + 0.758 =»= 0.028, 

 + 0.852 ± 0.018, and + 0.610 ± O-Oil, respectively. Correlations are calculated for yields 

 of corn on the same plots in the years 1895, 1896, and 1897, at the Illinois Experiment Station. 

 Yields of corn in the unfavorable year 1895 correlated with yields of the same plots for the 

 favorable years 1896 and 1897 are — 0.354 ± 0.054 and - 0.221 ± 0.059, respectively. 

 Yields for the two favorable years 1896 and 1897 show a correlation of + 0.818 ± 0.020. — 

 Data were taken from the records of the Field Station of the Office of Western Irrigation 

 Agriculture at Huntley, Montana. Yields were used from 46 plots in a uniform cropping 

 experiment covering the period 1909 to 1919 inclusive. The crops involved in the rotation were 

 sugar beets, alfalfa, corn for grain, oats, corn for silage, and barley. In some cases the 

 plots were subdivided into half and quarter plots. Correlations are calculated between yields 

 of the same plots for different years. Of the 152 correlations calculated for whole plots, 133 

 are positive and 19 are negative. The average value of the positive correlations is -1- 0.3346 

 as compared with — 0.1475 for the negative correlations. For the 152 constants the average 

 value is + 0.2743. — The data from half and quarter plots substantiate the results for whole 

 plots. The results show conclusively that in general plots which yield more in one year 

 will yield more in other years. — Some evidence is given to show the effect of the yield of a 

 particular crop in the rotation upon the yields of subsequent crops. Plots giving high yields 

 of sugar beets in 1911 showed low yields of alfalfa in 1912. Alfalfa exerted a definite residual 

 influence upon subsequent crops. The effect was not fully shown until the 2nd year after 

 the alfalfa field was plowed, and decreased until little if any effect was shown in the 4th 

 year. Due to the residual effect of alfalfa and its early introduction into the rotation it is 

 impossible to determine to what extent the correlations between the yields of alfalfa and 

 the yields of other crops are due to the variation from plot to plot of the amount of nitro- 

 gen fixed by the alfalfa and to what extent due to the original heterogeneity of the experi- 

 mental plots. — Fred Griffee. 



1088. Hensen, V. Die Mutation und was sie iiber die Entstehung neuer Arten lehrt. 

 [Mutation and what it teaches about the origin of new species.] Schrift. Naturw'iss. Ver. 

 Schleswig-Holstein [Kiel] 17: 1-12. 2 fig. 1920. 



1089. Hensen, V. Der Vorgang bei der Mutation. [The process in mutation.] Schrift. 

 Naturwiss. Ver. Schleswig-Holstein [Kiel] 17: 190-191. 1920. 



1090. Honing, J. A. Erfelijkheidsleer zonder Evolutie theorien. Rede uitgesproken bij 

 de aanvaading van het Hoogleeraarsambt aan de Landbouwhooge school te Wageningen 23 Nov. 

 1920. [Genetics vs. theories of evolution. Lecture given on assuming the office of Professor 

 at the Agricultural .High School of Wageningen, Nov. 23, 1920.] H. Veenman: Wageningen, 

 1920. 



1091. Honing, J. A. Selectie-Proeven met Deli-Tabak. [Selection studies with Deli 

 tobacco.] Teysmannia 30: 1-11. 2 -pi. 1919. — To increase production without extending 

 acreage, strains may be sought which produce a large number of leaves per plant. For cigar 

 wrapper-leaf, however, quality is more important than quantity production, hence high leaf 

 number is not given first consideration in selection work. Most types of high leaf number 

 are hybrids which continue to segregate indefinitely. Wrapper leaf should be of sufficient 



