arranged in pairs each bearing at the apex a small gland-like 

 more or less imperfect anther-cell containing aggregated pollen- 

 grains, sometimes few, sometimes empty (?), without any viscid 

 gland ; stigma sometimes terminal undivided cushioned hairy, or 

 stigmas 2, of which one is terminal, the other expanded into wings 

 on either side of the posterior pair of rostellary arms ; capsule 

 clavate, seeds with a i^erfect embryo. 



Described and drawn from living plants of both forms sent by Miss 

 C. B. Newdigate as above, and also in part from living plants of the 

 cleistogamous form only, sent by Mr. Glass from near Grahamstown. 

 Colour of the flowers sulphur-yellow ; the whole plant dries a pale 

 straw colour. Of the normal form, which seems to be somewhat rare, 

 I received first a one-flowered plant which is figured, and subsequently 

 two taller but withered 3-fl. specimens. The cleistogamous form was 

 found abundantly at all the stations quoted. The species is allied to 

 Pt. catholicum, differing by the larger, more oblong, lacerate lip, in the 

 shape of the appendage of the lip, and in the incised petals. Its chief 

 interest lies in its exhibition of a cleistogamic form so different both in 

 its perianth and in its column from those of the normal form that it 

 might, if it stood alone, be taken as the type of a distinct genus. 

 Cleistogamy is not unknown in the order ; but it is rare. Mr. Kolfe 

 has been good enough to refer me to the following papers and 

 instances : Dendrohium crepidatum, Lindley, noted by him in the 

 Kew Herbarium; — Eeichenbach fil., on Maxillaria rufescens, Lindley, 

 in Eefngiuni Botanicum, vol. ii, sub t. 133 (1882) ; — H. 0. Forbes, On 

 Self-fertilisation in Tropical Orchids, in Linn. Soc. Journ. (Bat.), vol. 

 xxi, 2^. 547 (1885), where a case of cleistogamy in Chrysogiossum sp. 

 is recorded and iigured in Plate 17, figs. G-11 ; — H. N. Eidley, Notes 

 on Self-fertilisation and Cleistogamy in Orchids, in [jinn. Soc. Journ. 

 (Bot.) vol. xxiv, ]). 390 (1888), recording cases of cleistogamy in 

 Triclwpilia fragrans , Lindley (Plate 16, figs. 1-9), and in Dendrohium 

 chryseum, Eolfe. None of these cases appears to throw any light upon 

 the present one. The erect processes shown in figs. 12 and 13 appear 

 to be homologous with the organs regarded as arms of the rostellum 

 in the normal form (see Orchids of the Cape Peninsida, p. 183, and 

 Plate 12, fig. 7) ; or they may perhaps be a development of the con- 

 nective of the anther. In this genus and in Ceratandra the anther- 

 cells are usually adnate to these arms, and this may also be seen in 

 Plate 48, figs. 6 and 7, of the present work. The anther-cells at the 

 apices of these arms in the cleistogamous form are of variable deve- 

 lopment. Sometimes they are nearly perfect, the pollinia, however, 

 having no proper caudicle and viscid gland ; in some cases the pollen 

 is reduced to a few isolated granules, and even sometimes (though 

 probably not in all of one flower) is wanting altogether. The presence 

 of four anther-cells, equivalent to two anthers, would be very curious, 

 and needs confirmation. No opened or punctured flower has, as yet, 

 been observed ; no pollen has in any case been found upon the stigma ; 

 yet the ovaries have swollen, producing abundance of seed, and Dr. 

 Schoenland, who examined them, found the seeds to contain a perfect 

 embryo. The question remains. How is the fertilisation eft'ected ? 

 This, with some other points in respect of this remarkable species, 

 must await for its elucidation a supply of fresh material, which Dr. 

 Schoenland, who has kindly assisted me in this matter, hopes to obtain 

 during the ensuing season. 



