1118 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



Salsify Hybrids. — The hybrid plants Tragopogon pratensis x T. 

 porrifolius reported on in U.S.A. Exp. Stn. Rep., New Jersey 1900, 

 produced seed sparingly. There are two types in the crossed plants 

 raised from these seeds as regards the colour of the inflorescence, (1) with 

 both ray- and disc-florets coloured alike, darker red, red violet, and (2) 

 those with ray-florets dark, while disc-florets are yellow. There are no 

 intermediates. In the crosses the violet has diminished, while the red 

 has greatly increased, and the yellow is scarcely in evidence in one of the 

 types. 



Other Experiments in Hybridising. — Martynia Louisiana, Mill, was 

 crossed with M. lutea ; an acceptable pickle may be thus produced. An 

 attempt was made to cross Tecoma radicans with Martynia, but un- 

 successfully, as was an attempt to cross Linum grandiflorum with 

 L. usitatissimum, L. In the latter case several full-sized seed-vessels 

 were produced, but they contained no seed. — F. J. C. 



Evils of Premature Pollination. 



Pollination, Injurious Effects of Premature. By Ch. P. Hartley 



(U.S.A. Dep. Agr. Bur. PI. Ind., Bull. No. 22, with 4 plates of photos ; 

 1902). — The object of the paper is to show that "a premature pollination 

 is destructive to the flowers of some plants, preventing them from forming 

 seeds. With Tobacco the growth of the pollen-tubes into ovaries before 

 the ovules are mature enough for fertilisation results in an injury which 

 causes the flowers to pale." 



Numerous experiments are described in detail, giving as the extreme 

 difference 20 flowers pollinated two days before opening ; no seed-pods set. 



20 flowers, do. v> day ; 19 seed-pods were set, or 95 per cent. 



" Dividing the six series of experiments into two groups, namely those 

 pollinated more than one day before opening, and those pollinated one 

 day or less than one day before opening, they show that the former set 

 4 per cent, of seed-pods which contained no germinative seeds, while the 

 latter set 86 per cent, of seed-pods which contained germinative seeds." 



A microscopic examination showed the pollen-tubes within the ovary, 

 but none can penetrate the micropyle in the immature ovule, and "in just 

 what way the flower has been destroyed has not been determined." 



Experiments with Datura Tatula showed similar results to those with 

 Tobacco. Other experiments were tried with flowers of Cotton-plant and 

 Oranges, but with less decisive results. With the Orange, " it was demon- 

 strated that fruits containing good well-developed seeds will result from 

 flowers pollinated nine days before they could naturally have received 

 pollen." — G. H. 



Pollen in Conifers. 



Pollen-sac Dehiscence in certain Conifers. By K. Goebel 

 (Flora, vol. xci. 1902, pp. 236 255 ; 13 figs.). — The mode of dehiscence 

 is in direct relation to the position of the pollen-sacs at anthesis, so as to 

 facilitate the ready escape of the dusty pollen. Where the sacs are 

 vertical or down-turned, the split is longitudinal (Finns, Picea). Where 

 the male cones are down-turned the slit is oblique, and expands below 

 into a spout (Larix, Abies). In Tax as the epidermis which forms the 



