532 MR. C. BUCKNALL : A REVISION 
that, in plants which he cultivated, they were white. It is remarkable that 
the plant appears to be always sterile. Schultz only found very small 
abortive fruits, and in all the specimens which I have seen, some of them 
well past flowering, the fruits were in the same condition. 
I must here give my reasons for concluding that this species, being a 
branched plant of the Ramosa division, cannot be, as Schultz supposed, 
A. mediterraneum, Koch, because, as I hope to prove, the latter is an 
unbranched plant allied to S. tuberosum. Unfortunately, Koch's type cannot 
be found, and we therefore have to rely on his deseription, which is as 
follows :—*“ S. mediterraneum rhizomate ...., caule folioso, apice bifido, 
foliis ovatis, inferioribus in petiolum basi subdilatata amplexicaulem contractis, 
supremis elliptico-lanceolatis subdecurrentibus, corollis subinfundibuliformi- 
eylindrieis, limbo quinquedentato, dentibus erectis limbo quadruplo brevi- 
oribus, antheris filamento duplo longioribus, fornicibus inclusis. Antherse 
basi cordatee ut in omnibus, sed auricule in hae specie mucronulate. Differt 
a S. tuberoso, foliis caulinis inferioribus cæteris majoribus petiolo basi dilatato 
amplexicaulibus et summis tantum decurrentibus, corollis dimidio minoribus, 
dentibus ipsarum erectis et antheris utrinque apieulatis. — Radieem et 
fructum nondum vidi." Koch states that he at first confused this with 
S. bulbosum, Schimp., and as he also compares it with S. tuberosum, Linn., 
it is evident that his plant bore a general resemblance to those species. An 
analysis of the description will show that it does in fact belong to the 
Tuberosa, and not, like S. floribundum, to the Ramosa. 
Omitting the characters which are common to both groups, and those of 
the root, which was not represented in Koch’s specimen, the points to be 
noticed are : the stem with bifid apex, which implies that it was otherwise 
unbranched ; the petiole with dilated and clasping base, which is charac- 
teristic of some of the forms of the Tuberosa, and is more or less obvious in 
the lower leaves of S. tuberosum itself ; the apiculate base of the anthers, a 
character which is rarely found in the Ramosa, but is always present in the 
Tuberosa. These characters all indicate that Koch’s plant belongs to the 
трба, and comparing it with Shuttleworth’s plant we see that this 
differs from it in the strong branched stem, the petioles not dilated and 
clasping the stem, and the anthers, with obtuse base, scarcely longer than the 
filaments. Moreover, S. floribundum possesses conspicuous characters of its 
own which Koch could scarcely have overlooked if they had been present 
in his plant, viz., the very numerous shortly-stalked flowers, and the cam- 
panulate calyx divided only to one-third into triangular teeth, forming a 
characteristic inflorescence very different from that of the plants with which 
he compares his А. mediterraneum. 
From these considerations it will, I think, be admitted that the two plants 
are perfectly distinct, and that I am justified in restoring Shuttleworth’s 
name to his plant, 
