AGRICULTURAL BOTANY. 521 



effect of limited sbade, varietal influence upon the length of hypocotyl anil first 

 internode, the intluenco of pod position of seeds upon length of hypocotyl and 

 vigor of the plant in soy beans and pea beans, and a study of the relation of 

 sex to the length of hjTocotyl in hemp. Notes are also given on the rise of the 

 zone of geotropic response in seedlings and of an experiment in the selection of 

 Wilson soy beans. 



A peculiar negative correlation in CEInothera hybrids, G. H. Shtjll (Jour. 

 GcneHcs, If (19 U,), No. 1, pp. 83-102, pis. 2, fig. i ) .—Referring to difficulties in 

 applying to CEnothera the rules of genetic behavior demonstrable in other 

 groups of organisms (suggesting a hereditary mechanism in this genus funda- 

 mentally different from that which distributes the Mendelian unit characters), 

 and noting an apparent inadequacy of the data previously recorded for dis- 

 covery of the essential features of this mechanism, the author presents and 

 discusses data obtained by himself in studying the progeny of three rosettes 

 from unguarded but probably pure bred seeds of CE. rutricalyx since 1912, both 

 in pure strain and in crossing with other forms. 



It is held that three conclusions arrived at by Gates (E. S. R., 25, p. 327), 

 regarding the origin and genetic nature of the (E. rnhricalyx character are erro- 

 neous, namely, that the character represents a purely quantitative difference 

 from CE. rubrinervis, that it differs from the latter si>ecies in a single mono- 

 hybrid Mendelian unit, and that the nature of a character itself, instead of the 

 nature of the inheriting mechanism to which it is related, determines the 

 manner of inheritance of that character. 



Some fundamental morphological objections to the mutation theory of De 

 Vries, E. C. Jeffrfa' (Aincr. Nat., 4<J (1915), ^-o. 577, pp. 5-21, fif/s. 7).— As a 

 result of his studies the author concludes that the Onagracese are largely char- 

 acterized by hybrid contamination in nature. This holds particularly for 

 Q^noihera lamarckiana and other species of this genus, which serve as the basis 

 of the mutation hyi:)othesis of De Vries. The species of CEnothera, the author 

 says, are to a large extent, if not wholly, crypthybrids, and in his investigations 

 the objection raised bj^ Bateson to the genetical purity of GE7. lamarckiana was 

 confirmed. He considers hybridism the best explanation for the peculiar con- 

 duct of CE. lamarckiana as well as other species of this genus in cultures. 



The explanation of an apparent exception to Mendel's law of segregation, 

 TiNK Tammes (K. Akad. Wetensch. Amsterdam, Versl. Wis en Natnurk. Afdeel., 

 22 (1913-14), pt. 2, pp. 846-857; also in ditto, Proc. Sect. Set., 16 (1913-14), pt. 

 2. pp. 1021-1031; and Rec. Trav. Bot. N^erland., 11 (1914), A'O. 1, pp. 54-69).— 

 The author, from experiments in crossing white and blue flowering varieties 

 of flax, found in the second and following generations white and blue indi- 

 viduals that were not in agreement with the numbers expected in accordance 

 with Mendel's law of segregation. In all cases there was a deficiencj' in white 

 flowering plants. This is believed to have ai'isen from two causes, (1) the F2 

 genex-atiou and the heterozygotes forming too small a numljer of seeds to yield 

 white flowering plants, and (2) the low germinating power of the seed. 



The infl.uence of position in the pod upon the weight of the bean seed, 

 J. A. Harris (Amer. Nat., 49 (1915), No. 577, pp. 44-47, figs. 3).— Attention 

 having been called to the significance of position in the pod as a factor in de- 

 termining the weight of the bean seed, the author gives the results of a series 

 of quantitative determinations of the relationship, basing his figures largely on 

 data part of which have been previously given (E. S. R., 30, p. 433). 



He claims that the percentage of ovules which develop into seeds increases 

 from the base toward the stigmatic end of the pod. In small r>ods the rate of 

 increase may be fairly regular, but in larger pods it falls off toward the stig- 



