HYBBIDS OF ORCHIS PURPUBELLA 35 



Our pa])er (Joavn. 1920, iG^t- 17u;, and the close connection of the 

 two forms is recognised. We found at Arran jjlentv of both foruis, 

 whieh there gi'ow together. The tiowers are of precisely the same 

 tyjje, the differences being that Cpurpurella is dwarf, w-ith minutelr 

 spotted leaves, whilst O. jjulchella is much taller, stout or slender, 

 and with unspotted leaves. More work needs to be done on the 

 forms ; but we are inclined to think that we have here a case of 

 linked species, differing only by the unit-characters for " dwarfnoss " 

 and " leaf-spots." As far as present obser%'ations go they aiv mostlv 

 found apai-t, though at Arran they are found constantly together. 



Those who do not recognise O. Jatifolia as a true British species 

 will quarrel with our X O. insignis. It may be said that the type is 

 very distinct from that of the other two, and needs to be named and 

 described, even if its exact origin be questioned. The number of 

 individuals is relatively large. It is interesting to note that, though 

 in Ai-ran O. prcEtermissa is entirely absent and O. pidcheUa takes its 

 place, examples of O. latifoUa are to be found exactly like those in 

 areas where there is no puIcJieUa. That an identical type should 

 arise from two such dift'eivnt parents seems most unlikely. 



The other figure here reproduced (PI. 561 B) fi-om the OrcJiirl 

 Review is Orchigymnadenia Evansii {Gyjtinadenia conopseay.0. 

 maculata subsp. ericeforum Linton : X O. Evansii Druce in Keport 

 Bot. Exch. Club, 1906, p. 199). The spike was pale lilac in colour 

 with three minute dots in rows on each side of the lip. The flower 

 was scented. The gi-eat length of the spur is well seen in the 

 photogi-aph. The plant was about 2'5 dm. high, with long, narrow, 

 rather fleshy, unsixttted leaves. A solitary specimen was found in 

 Arran last July, the finest example of this hybrid we have seen. 

 We saw a dozen or more examples in Wales, all charming'y orraceful 

 plants, varying a good deal in colour and markings, but all with verv 

 long spm- and scented flowers. 



P.S. — In the Orchid Eevieu: for December, we published a photo- 

 graph of form A of O. purpureJla, here reproduced ( t. 562). with a brief 

 note on the species. Two points raised in our paper in this Journal can 

 be fui-ther developed. Form B is xery closely connected with O. prce- 

 fermissa v. pidcheUa Druce. In Arran they are found together. It 

 can now be quite definitely stated that O. purpureUa is not the same 

 as O. cruenta O. F. Mueller, which is much nearer to O. incarnata L. 

 The simplest way of stating the relationship of the various forms is 

 to say that O. cruenta and 0. purpureUa are dwarf spotted-leaved 

 variants of O. incarnata and O. pulchella respectively: but the leaf- 

 markings of O. cruenta are of a very different type from those of 

 O. purpureUa. In view of this we suggest that it would be advisable 

 to separate O. puIcheUa from O. prceterniissa, hrom. which, in the 

 floral characters, it diffei-s a good deal. It is now certain that pre- 

 vious recoi-ds of O. cruenta for Britain are incorrect ; the plants 

 found must be assigned to O. purpureUa. 



r 2 



