428 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



was not increased in the fifth generation, nor were the plants produced 

 from "blind" wood less floriferous in the fifth generation. Where 

 "bloom" rather than stock plants is the end sought, cuttings taken from 

 " flowering " wood are far superior to those taken from " blind " wood. 



Improvement of Oats by Breeding. By J. B. Norton, of 

 Washington. 



This paper contains a description of the work recently undertaken by 

 the United States Department of Agriculture in the breeding of Oats to 

 secure rust resistance, hardiness, and increased yield. With regard to the 

 cross-fertilisation of the Oat, pollination is best done from 1 to 3.30 p.m., 

 as it is hard to find ripe anthers before this time, and later than this most 

 of the pollen has escaped. Flowers emasculated one day are pollinated 

 in the afternoon of the following day. In dry hot weather only about 

 5 per cent, of successful crosses were made, but in cool moist weather 

 75 per cent, were obtained. Natural crosses in Oats appear to be rare, 

 indicating that Oats are nearly always self-fertilised. The individual 

 plant is taken as the basis of selection : this is found absolutely necessary 

 for good work on account of the great individual differences found in the 

 same race. 



On Breeding Florists' Flowers. By E. G. Hill, of Indiana. 



Mr. Hill deals in a general way with the breeding of Roses, Carnations, 

 Chrysanthemums, and Begonias. In Roses, out of several thousand 

 seedlings, less than two dozen were of permanent value, and a few only 

 were put into commerce. The ripening of the seeds was the most difficult 

 task in Rose breeding. In Chrysanthemums the large double flowers gave 

 little seed, but these produced large full flowers, On the contrary in 

 Roses and Carnations, when both parents were full-petalled, a large pro- 

 portion of singles and semi-doubles were produced. In Begonias Mr. Hill 

 obtained a whorl-leaved variety of ' Rex ' by crossing the ordinary ' Rex ' 

 with the whorled variety, ' Countess Erdody ' ; about thirty of these were 

 raised having the 1 Rex ' markings and colouring, with a distinct single 

 or double whorl of the leaf. 



A Medley of Pumpkins. By L. H. Bailey, of New York. 



Professor Bailey records some of his experiments in crossing 

 cucurbitaceous plants from 1887-1897. The original object of the investi- 

 gation was to determine the question of the immediate influence of pollen 

 on the fruit (Xenia). More than 1,000 hand crosses were made, and in no 

 case was there a trace of Xenia. The popular idea that " Cucumbers spoil 

 Melons " is therefore not confirmed. Most of the crosses were between 

 races of Cucurbita Pepo. In the first, second, and third generations there 

 was the greatest possible diversity in the progeny. More than 1,000 kinds 

 of fruit were produced, and in one season eight acres of ground were 

 required to grow the plants. The magnitude of the results prevented 

 detailed publication, as no underlying principles were discovered. New 

 characters appeared that were wholly lacking in either parent, e.g. unusual 

 colours, shapes, and wartiness of fruits. 



