242 



DR J. M. MACEAELANE ON THE 



/. Masdevallia 

 amabilis. 



.' Masdevallia 

 Chelsoni. 



; Masdevallia 

 Veitchiana. 



We thus have overwhelming evidence in favour of the hybrid nature of Bryanthus, and 

 a true index of its parentage. As already stated, the report very early gained currency 

 that Menziesia ccerulea — not empetriformis — was one parent, and this statement has been 

 perpetuated in standard works on the subject. The hybridiser clearly stated and has 

 reasserted to me that M. empetriformis, var. Drummondi, was that used by him. What 

 evidence, then, it may be asked, can be adduced in favour of the latter over the former ? 

 As Mr Lindsay has well pointed out, the hybrid has a growth-vigour which would 

 scarcely be expected from union of the dwarf Rhododendron Chamcecistus with the 

 equally low-growing M. ccerulea, even if we allow for that luxuriance which hybrids 

 occasionally show. But additional evidence is needed, and this, I think, is furnished by 

 the calyx and corolla. The external surface of the calyx of M. ccerulea is densely 

 studded with glandular hairs, which are scarce in M. empetriformis. The corolla is 

 ovate-urceolate in the former, and has a coating of gland hairs over its outer surface. 

 The corolla of the latter is tubular, constricted above, and has transverse plications ; its 

 outer surface is also quite glabrous. At least two considerations militate in favour of the 

 variety "Drummondi" having been used. These are the colour of the flower, which has 

 given to the hybrid its clear delicate pink hue ; also the time of flowering, for 

 records of the yearly flowering periods at the Royal Botanic Garden place M. empetri- 

 formis earliest, Rhododendron a few days later, and Bryanthus later still. But the 

 variety Drummondi was this year later in flowering by sixteen days than the ordinary 

 species. This, however, would accord well with the flowering period of the hybrid and 

 of the other parent. Our evidence, therefore, is wholly favourable to the accuracy of 

 Mr Cunningham's statement. 



I would further venture to assert, with considerable confidence, that when a cross is 

 effected between Rhododendron and Menziesia ccerulea the progeny will be low-growing 

 plants, with dull pink flowers ; that the calyx externally will be more glandular than in 

 Bryanthus, that the corolla will be tubular and non-plicate, and that a considerable 

 number of glandular hairs will be found over its outer surface. 



(h) Masdevallia Chelsoni, x . 



This hybrid was raised by Mr Seden, a well-known hybridiser on Mr Veitch's staff. 

 It was described in the Gardeners' Chronicle for 1880 (pp. 501, 554). M. amabilis, 

 the seed parent, is figured and described in Bonplandia and Illustrations Horticole, 

 and occurs wild in Northern South America. The pollen parent is found on the Cor- 

 dilleras of Peru, was first discovered by Pearce, flowered by Messrs Veitch, and figured 

 by Sir Joseph Hooker in the Botanical Magazine, No. 5739. 



I regard the parents as closely related species, on account of their naked-eye appearance 

 and microscopic structure. We shall see, however, that there are several important points 

 of difference, especially in histological details. Like other Masdevallias, they have a tufted 

 habit, form a dense mass of roots whose upper parts are green, have short cylindric 



