540 MR. C. BUCKNALL: A REVISION 
The var. angustifolium, Kern. is a very distinct looking plant, of which 
I have only seen the specimens cited. In these the primordial leaves are 
very small and narrow, measuring with the petiole 3 em. x 0*3 em.; the 
largest stem-leaves are 10 em. x 1-4 em., the flowers 15 mm. long, the calyx 
8 mm. long, with nearly filiform segments only 0:5 mm. wide at the base. 
In Dr. Woloszezak's plants, eultivated at Kaschau, the leaves are rather 
broader in proportion to their length. The Trenesin plants referred to 
S. angustifolium have small leaves which are mostly similar in [shape and 
proportions to those of typical S. tuberosum, but in some cases they show a 
tendency to become more elongated. These cannot be regarded as repre- 
senting Kerner’s plant. 
S. foliosum, Rehm. differs from S. tuberosum, according to the description, 
in the radical leaves being very numerous, and in the corolla being twice ав. 
broad. Specimens from woods on the banks of the Dniester at Czernelica, 
the locus classicus, are nearly past flowering, and therefore not in a condition 
to show any distinctive characters which they may have possessed. The 
stem is about 3°5 dem. high, nearly naked for more than half its length with 
leaves crowded towards the top, and the few remaining flowers are small. 
Otherwise there is nothing to distinguish these specimens from ©. tuberosum. 
The plants of the Tuberosum section are more nearly related than those of 
most of the other sections of the genus, and opinions have differed greatly as 
to their rank. Caruel states that, having studied these plants in the neigh- 
bourhood of Florence, where all the forms grow promiscuously together, he 
has formed the opinion (Fl. Ital. vi. p. 879) that S. mediterraneum, Gussonet, 
bulbosum, and Zeyheri are all different forms or states of S. tuberosum, and 
that this species varies in the following manner :—The root is thickened 
in different ways, sometimes for a considerable distanee as in Reich. Icon. 
t. 103, sometimes becoming large and nodose as in Jacq. t. 63, and sometimes 
it is thickened at intervals into lateral, sessile or stipitate tubercles as in 
Bischoffs figure of S. filipendulum (S. bulbosum, chimp.) in ‘Flora,’ ix. 
p. 561; the flowers vary in size, sometimes the entire flower, sometimes 
only the corolla being smaller, and in the latter case the corolla-scales and 
style ure necessarily exserted. 
But it should be borne in mind, firstly, that where 5, tuberosum and 
S. bulbosum grow together, the circumstances are not favourable for forming 
an opinion as to the relationship of the members of this group, on account of 
the probability of hybridization taking place between them and producing a 
series of forms which may be confused with independent species. Secondly, 
tnat other characters must be taken into consideration besides those of the 
root and flowers on which Caruel relies. Thirdly, that so far as is known, 
the plants mentioned above do not appear to be hybrids, but stable forms 
occupying separate geographical areas. In the case of S. tuberosum and 
