370 RHINANTHACEAE 



the Bot. Exchange Club. It is an exceptionally narrow-leaved, slender 

 plant with handsome flowers, the corolla whitish-yellow with large 

 and conspicuous palate of deep orange colour. The chief difference is 

 in the spur, which in vulgaris is described by Syme as nearly straight, 

 and by Hooker as parallel, and longer than the tube. In this variety 

 the spur is so much bent as to be almost at right angles to the tube. 

 The upper petals are so much reflexed that when looked at from 

 the front they present only their edges to view. The pedicels are 

 covered with gland-tipped hairs. W, A. Bromfield alluded to a form 

 with milk-white corolla and palate deep orange, as var. 7, in the Phyt. 

 (1849) 627. A form of otherwise normal L. vulgaris, which had three 

 spurt-, was seen at Maidenhead. 



The qnestion arises whether the variability of L. vulgaris (and other species) 

 may have been caused by the pollination by a second species in one of its 

 ancestors, so that the stability of the species has been disturbed ? This may 

 have taken place, not necessarily in the generation immediately preceding, 

 but in a more remote ancestor, so that the hybrid influence had been 

 gradually neutralized by successive pollination with a similar species. 

 A hybrid is almost always handicapped by the more numerous individuals 

 of one or the other parent, so that the chances are immensely against the 

 pollination by a similar hybrid form. Hence in fertile hybrids the influence 

 of one or other of the parents is gradually eradicated. My point is that 

 after the traces of cross-pollination have disappeared, it does not follow that 

 an equally stable species remains as that which existed before the cross- 

 pollination took place. I think it will be found that the plant is more 

 distinctly liable to variation, not necessarily in the direction of its remote 

 ancestor, but in other wajs. To remote hybridization I think must be 

 attributed the extreme variability exhibited by certain genera, such as 

 Eiibus, Rosa, Salix, Primus, Epilobium, &c. ; these genera, as we know, being 

 especially prone to hybridization. 



*Ii. PURPUREA, Miller, 1. c, n. 5 (1768). Purple Toad-flax. 



Antirrhinum purpureum, Linn. Sp. PI. 613. 



Comp. Cyb. Br. 54.1. Syme, E. B. vi. 138, t. 960. Nyman, 539. Fl. Oxf. 212. 

 Alien. Old walls. Bather rare, and in most cases a mere garden escape, 



but readily establishing itself if undisturbed. P. June-September. 

 First found in Berkshire by the author in 1880. 



1. Isis. Wytham. 2. Ock. Marcham, on the village wall. Ken- 



nington. Hanney. 4. Kennet. Southcote. Beading Abbey. 



5. Loddon. Maidenhead. 



L. purpurea is reported from Oxfordshire, Bucks, Hants, and Wilts. 

 li. viscida, Moench, Meth. 524 (1794). Small Toad-flax. 



L. minor, Desf. FI. Atl. ii. 46 (1799). Antirrhinum minus, Linn. Sp. 

 PI. 617 (1753). Chaenarrhinum minus, Lange, in Willk. and Lange, 

 Prod. Fl. Hisp. ii. 579. 



Top. Bot. 297. Syme, E. B. vi. 143, t. 966. Nyman, 542. Fl. Oxf. 214. 



Native or colonist. Agrestal. Cultivated ground, waste places. Rather 

 local. Particularly fond of railway ballast, and frequent along 

 the permanent way of the Great Western Railway. A. May- 

 September. 



