262 



P. F. YEO 



it now inhabits. If this is so, one must assume that in some areas E. anglica was once 

 present and has now become extinct, probably owing to the elimination of its habitat. 



Diploid-Tetraploid Hybridisation as a Source of New Species 



E. vigursii is in effect a new species, assuming the hypothesis offered to account for 

 its origin is correct. This suggests that other species may have been formed in the same 

 way. The two forms most likely to have arisen in this way would appear to be E. hrevipila 

 Burnat & Gremli var. notata Pugsl. and var. reayensis PugsL, since they combine characters 

 typical of tetraploid species with the long-glandular indumentum of the diploids. The 

 former occurs in central Scotland and is best known around Loch Tay. Here there also 

 occurs a diploid species, E. hirtella. E. hrevipila var. reayensis occurs along part of the 

 north coast of Scotland, in Caithness and West Sutherland. No diploid species has, 

 however, been reported north of the Loch Tay area. These two varieties of E. hrevipila 

 were investigated in 1953 and both were found to be tetraploid, as already reported (Yeo, 

 1954), 



It is conceivable that E. rivularis may have arisen from E. rostkoviana and E, curta 

 (Ft.) Wettst. var. rupestris PugsL 



There are several other cases of apparent hybrids between diploid and tetraploid 

 species, and these will be considered in the next two sections. 



Further forms analogous to E. vigursii 



Under this heading hybrids in which the habit of the diploid is modified will be 

 dealt with. When these have been seen in the field the habitat has also been found to 

 be modified. When they have not it is nevertheless to be expected that in fact the habitat 

 is modified. All such forms are thus thought to be strictly analogous to E. vigursii. 



(1) E. anglica X nemorosa 



Pugsley (1930) described a form from Piltdown Common, E. Sussex, v.c. 14, as E. 

 anglica X pseudokerneri Pugsl. A modified form of E. anglica has also been collected 

 at Lane End Common, two and a half miles west of Piltdown {Milne-Redhead & Summer- 

 hayes, No. 1085, 1937, in Herb. Kew). The herbarium specimens from both these 

 places are long-glandular, like E. anglica, but rather tall and large-flowered, and differing 

 from E. anglica in leaf shape. 



On visiting these two places, I was unable to find pure E. anglica, but at Piltdown 

 (grid ref. 51/444224-446226) two forms of the glandular plant were found. One was 

 only slightly modified; it had green leaves as in E. anglica, but the leaves were narrower 

 and the internodes longer. It grew in grass of moderate length. The other form was 

 much more modified, being taller and stouter-stemmed, and having a good deal of 

 anthocyanin in the vegetative parts. It looked much like the Roborough form of E. 

 vigursii, but the flowers were white with the upper lip lilac. This form grew in a grass- 

 heath community, and with it was found a small colony of a form of E. nemorosa. This 

 form was rather small-leaved and usually had lilac flowers; it is possibly var. transiens 

 Pugsl. The same form was found at Lane End Common and a specimen from there 

 is shown in Plate 17, fig. 2, No. 5. 



At Lane End Common (grid ref. c. 51/404223), E. nemorosa (Plate 17, fig. 2, No. 5) was 

 much commoner than at Piltdown, growing in short turf along the edges of the paths. 

 The apparent hybrid showed variation according to habitat, tending towards E. anglica 



