160 GENETICS [BoT. Absts., Vol. X, 



1049. Elderton, Ethel M. [Rev. of: A study of women delinquents in New York State 

 by Mabel R. Fernald, Mary H. S. Hayes and Almena Dawley with a statistical chapter by 

 Beardsley Ruml and a preface by Katherine Bement Davis. Century Co.: 1921.] Biometrika 

 13: 305-308. 1921. — This is said to be the first use of modern statistical methods in a study of 

 women delinquents. Data are confined to mental capacity and main facts of personal and 

 environmental history, as adequate physical and medical facts could not be obtained. The 

 groups studied were: (1) New York State Reformatory women between ages of 16 and 30 

 convicted of felonies and misdemeanors; (2) Auburn State Prison felons; (3) Magdalen Home 

 sex offenders; and (4) New York County Penitentiary; (5) workhouse, and (6) women's night- 

 court cases. Inebriates are omitted because of inability to obtain accurate information from 

 them, thus eliminating many mental defectives and introducing a serious source of error Vari- 

 ation in the length of period over which the subjects were examined is also a defect. Mental 

 condition was ascertained by Binet Simon 1911, Yerkes Bridges Point Scale, Stanford Revi- 

 sion Binet, Wooley Series, Individual standard tests of ability, and Educational tests; and by 

 social investigation of home, relatives, employers and other agencies in touch with offender. 

 Some correlation was found between age and numberof pre viou s convictions. Foreign white wo- 

 men make up a smaller percentage in each institutional group than in the population as a whole. 

 Offenses of the foreigners were more serious than those of the natives. A correlation of 0.31 

 was found between condition of home and age at first conviction. In 15.9 per cent of families 

 some other member had been convicted of crime. No correlation was demonstrated between 

 age upon leaving school and first conviction nor between number of convictions and grade 

 reached in school. More domestic workers were included in the group than in the general popu- 

 lation, and a lower standard of education prevailed than in other occupations. The younger 

 women had^ reached a higher grade of education than the older. No correlation was found 

 between earning capacity and number of convictions. Neither lack of education nor low and 

 irregular wages are responsible for delinquency. No relation was seen between habitual use 

 of alcohol, drugs, or cigarettes and intelligence. A correlation of 0.38 occurred between 

 delinquency and intelligence. Also, significant correlations were found between mental 

 capacity and type of occupation, between wage capacity and type of occupation, and between 

 mental capacity and earning capacity. — No control series was used for comparison with 

 normal population. Comparison with men criminals is futile because the offenses of male 

 convicts are not comparable. The data are considered inadequate for treatment of 

 hereditary aspects. — Miriam C. Gould. 



1050. Fernald, M. L. The geographic distribution of hybrids. Science 54: 73-74. 1921.— 

 Kerner is quoted on cases of assumed hybrids of Nuphar, Salvia, and Rhododendron which 

 spread beyond the limits of the parent species and are then indistinguishable from true species. 

 Similar cases were recorded by the writer in Rubus. — T. D. A . Cockerell. 



1051. Fleischmann, RtJDOLF. Beitrage zur Leinziichtung. [Contributions to flax breed- 

 ing.] Zeitschr. Pflanzenziicht. 8 : 26-43. 1921 . — Certain results in practical fiber-flax breeding 

 are detailed relative to variation, correlation, and inheritance of stem length. Breeding 

 material was taken from fields in 2 districts of Hungary. Branched portion of flax plant 

 showed greater variability than lower, unbranched, portion. The author concludes that 

 clearer, more accurate results were obtained by limiting the study to unbranched part of stem. 

 Intensity of inheritance was measured by coefficient of correlation of length of stem from one 

 generation to next. This coefficient was generally high but varied according to weather 

 conditions. (Two crops were raised each season.) Inheritance of stem length was as pro- 

 nounced with selected lines as in general, unselected stock. In series of selected lines showing 

 decrease in stem length, the standard deviation was found also to decrease coincident with 

 approach to a symmetrical type of curve. Skewness decreased as mean length diminished; 

 this did not hold true in mixed populations. Absolute decrease of stem length, brought 

 about by weather conditions, was relatively greater in selected lines of maximum stem length; 

 results of this character were less marked in unselected stocks. Hot, moist weather in 2nd 

 half of 1920 decidedly increased the coeflacient of variation. Stem length was not significantly 

 increased, but flowers and bolls were markedly increased and also basal branches. Selection 

 of stem length within (presumably) pure lines showed no significant influence. — L. R. Waldron. 



