87 
t. 4312), E. Mane Hanst. (Bot. Mag. t. 5195), E. Smeg it 
Hook. f. (Bot. Mag. t. 5925), E. fulgida, Hook. f. (Bot. Mag. t. 
6136), and FE. Fendleriana, Kuntze. 
This leaves the genera Physodeira, Hanst., Alsobia, Hanst., 
Skiophila, Hanst., Centrosolenia, Benth., Paradrymonia, Hanst., 
richodrymonia, Ocrst., and. Natitocalye, | nrg (which are 
included under Episcia in the Genera Plant to be accounted 
for. All these have ovules on the sitet pier re only of the 
placental lamellae, as in Alloplectus, but differ from that genus in 
h f the corolla and in being herbs. It will only be 
necessary to take into consideration Centrosolenia and Nautilocalyx 
on the present occasion : definitions of the remainder* are given b 
Hanstein in Linnaea, vol. xxvi. pp. 206, 207, and Oersted, 
Gesneraceae centro-americanae, p. 
Centrosolenia was founded by Bentham in Hook. Lond. Journ. 
Bot. vol. v. 1846, p. 362, on C. hirsuta, Benth., a creeping herb 
collected in British Guiana by Robert Schomburgk. Neither the 
placentation nor ihe anthers were mentioned in the original descrip- 
ion. A flower-bud from the type specimen was _ therefore 
examined by the writer, who found that the placental lamellae bore 
ovules on the inner surface only, and that the anthers were didy- 
mous, with a didymous connective. Episcia densa, C. H. Wright 
(Kew Bull. 1895, p. 17; Bot. Mag. t. 7481), has similar placenta- 
tion and anthers, ‘and is therefore transferred to the genus Centro- 
solenia as C. densa, Sprague, comb. nov. It may be noted here that 
several of the genera segregated from Episcia have ee es 
nthers. Those of Nautilocalyr have a much thickened oblo 
connective, bearing two parallel mussel-shaped thecae, wish 
dehisce introrsely. 
The genus Winitilocalys oe described in 1853 by Hanstein 
(Linnaea, vol. xxvi. pp. 206, 207), according to whom the name 
peared two years earlier in Linden, Cat. 1851, p.12. According to 
Hooker and Hanstein, Linden gave ‘the name Nautilocalyx hastatus 
to the type species, but this was rejected by Sir W. J. Hooker, 
=hé described it in 1852 under the name Centrosolenia bractescens 
(Bot. Mag. t. 4675). There is no evidence to show that Linden’s 
Catalogue of the year 1851 contained a botanical description of the 
genus or species. The first valid description of the species appears 
to ee been published in 1851, as Centrosolenia bracteata, Planch. 
Sethe es Serres, vol. vi. p. 322, cum ic.), and the first valid descrip- 
tion of the genus seems to have been published in 1853 (Linnaea, 
vol. xxvi. pp. 206, 207). The type species should accordingly bear 
‘the name 1 autilocalys bracteatus 
' An enumeration of the species of Nautilocalyz is appended. As 
many of the best specific characters are with culty observable 
in ied state, it has not seemed desirable to attempt the con- 
struction of a clavis, but diagnostic characters are given under the 
species as far as possible. N. bracteatus is placed at the head of the 
‘list on account of its peculiar characters, and the remaining species 
follow in chronological order. 
* Skiophila pulchella i is to be excluded from Shiophila, Hanstein transferred 
it to Tussacia in Linnaea, vol. xxxiv. p. 337. 
