140 Hill. — A Revision of the Geophilous Species of Peperomia. 
texture, and internal structure of the leaf are of great value. A few 
additional points of biological interest, with regard both to the seedling 
structure and to the development of the bulb or rhizome, have come to 
light during this investigation, an account of which will be given below. 
Seven new species are described in the course of this paper, and the 
descriptions of most of the known species have undergone considerable 
emendations. 
History. 
The first of the bulbous species to be described was the Peperomia 
umbilicata of Ruiz and Pavon 1 in 1798, from the neighbourhood of Lima in 
Peru ; it is a well-marked form, which does not appear to have been brought 
to Europe since Pavon’s time. The bulb or tuber is more or less placenti- 
form in shape, the roots spring from the sides and base, whilst the upper 
part is bare and crowned by the rosette of radical leaves. The next 
specimens to arrive in Europe were those collected by Humboldt in Mexico, 
which are figured and referred by Kunth 2 in 1815 to the same species as 
Pavon’s Peruvian plant. In the figure the roots appear to emerge more or 
less from a point on the upper surface of the bulb, which is described as 
being about the size of a pea. 
In other respects the descriptions of the two plants agree, and since 
ripe fruits were not obtained, and the characters of bulb and leaf were not 
very clearly noted, it is scarcely surprising that these two plants should 
have been considered to belong to the same species. Since this time 
many geophilous forms have been collected both in Mexico and in Peru, 
and all those with a distinct bulb have been referred to P. umbilicata 
apparently without any critical examination of the plants having been 
undertaken. 
Thus in the Prodromus, at least four well-marked and easily separable 
species are included under P. umbilicata , R. and P., and even Dahlstedt, 
though he has removed certain plants from this collection to form his new 
species, P. peruviana (Miq.), Dahlst. 3 , has failed to separate the Peruvian 
from the Mexican forms, with which they have been so long confused. 
Specimens belonging to two distinct species are also included under 
P . parvifolia, C. DC., both by De Candolle and Dahlstedt 4 . It is owing 
1 Ruiz and Pavon, FI. Peruv., i, p. 30, T. 45, Fig. 6. 
2 H. B. and K., Nov. Gen., i, p. 59, T. 15, Fig. 1. 
3 Dahlst., 1 . c., p. 32. 
4 DC. Prod., XVI, i, p. 393. Dahlst., 1 . c., p. 30. P. parvifolia , C. DC. was described in Seeman’s 
Journal of Botany of 1866, p. 133, from Pentland’s specimen (from 12,850 feet) at Kew and from 
a Pavon specimen in the Boissier Herbarium. In the Prodromus in 1869, p. 393, Mandon’s plant 
No. 1123 is added to the two just mentioned, and the diagnosis is amplified by a description of the 
ovary ; the spikes also are said to be dense-flowered. The Pavon plant, which appears to be similar 
to Mandon No. 1123, has no ripe fruits, but was probably the plant on which the original description 
was based, for Pentland’s plant is without doubt a specimen of P. peruviana, Dahlst. 
