840 EXPERIMENT STATION" RECORD. [Vol.40 



Influence of foreign pollen on the development of vanilla fruits, T. B. Mc- 

 Clelland (Jour. Agr. Research [U. 8.1, 16 (1919), No. 9, pp. 245-251, pis. 5).— 

 In connection with adaptation studies conducted with several species of vanilla 

 at the Porto Rico Experiment Station, various reciprocal crosses were made at 

 Mayaguez with a view to the development of valuable strains. It was observed 

 that the fruits of these hybrids were noticeably different from the others on 

 the same vines, hence numerous additional crosses were made to study this 

 phenomenon. Tabular data are given showing the girth measurements of fruits 

 resulting from reciprocal crosses between Vanilla plant folia and several other 

 species commonly grouped together and called " vanillon." Variations among 

 the hybrids are also shown in photographical illustrations. 



Where to either the V. planifolia or the vanillon stigma pollen of the other 

 has been applied a very decided modification in the form of the fruit has 

 resulted. The modification is in most instances so decided that these fruits ran 

 be distinguished from close-fertilized fruit 6 : at a glance. The author attributes 

 this phenomenon to variations in flower structure. The column of the vanillon 

 species is much longer than that of V. planifolia. exceeding the length of the 

 latter in some instances by as much as 60 to 70 per cent. 



"It seems quite reasonable to suppose, from the heavy fertilization of ovules 

 near the apex and sparse fertilization or entire absence of fertilization near 

 the base of the ovary when the vanillon stigma has boon pollinated with 

 V. planifolia pollen, that these pollen tubes are unable to reach or reach in 

 only limited numbers the ovules in the far end of the ovary, which are at a 

 considerably greater distance from the stigma than the farthest ovules of the 

 V. planifolia ovary. Even in its own ovary, the V. planifolia pollen causes a 

 much heavier fertilization near the apex than near the base. This inability of 

 Y. planifolia pollen tubes to reach the farthest ovules was particularly marked 

 when V. planifolia pollen was applied to V43, which is one of the largest 

 flowered of the vanillon varieties. 



"The vanillon pollen tubes, however, reach ovules in the T'. planifolia ovary 

 at a much shorter distance from the stigma than in their own flower. Many of 

 these first ovules which the V. planifolia pollen would fertilize are left unfer- 

 tilized by the vanillon pollen, the pollen tubes passing by to other ovules which 

 are nearer the normal distance from stigma to ovary in the vanillon flower, and 

 causing a much heavier fertilization in the base of the pod than would the 

 Y. planifolia pollen. 



"This might possibly indicate in this instance the necessity for a certain 

 maturity of development of the pollen tube before the ovule can be fertilized." 



Notes on geranium breeding. W. R. Ballard (Proc. Amcr. Soc. Hort. BcL, 

 15 (1918), pp. 62-65). — Some breeding experiments were started with the 

 geranium at the Maryland Station in 1907 with the view of improving it as a 

 bedding plant. Some 125 named varieties of the zonal group and a few plants 

 of the Lady Washington and scented-leaved sections were used in the work. 



Several hundred seedlings have been grown, but the greatest difficulty has 

 been encountered in the attempt to combine the disease-resistant character with 

 the flower characters which would make them of special value for bedding. 

 Because of the hybrid nature of most varieties considerable segregation takes 

 place in the F, generation. The results from one season's crop of seedlings In 

 relation to the form of the flower gave the following totals: Singles X singles 

 gave 84 singles and 6 doubles; singles X doubles gave 69 singles and 74 doubles; 

 doublesX doubles gave 40 singles and 30 doubles. 



In the course of the work it was observed that Hie neetar tube was not a 

 constant character in the zonal group, but was apparently correlated with the 



