FLORA OF SOUTH AFRICA. 81 
frutices mont. Tygerberg (III, D. a.) Maj. 1838. Hb. Krauss, 
propr. 
84. P. Kraussiana, nob.—Glabra, ramis strictis, erectis, 
lineatis, dense foliosis; stipulis lanceolato-setaceis, petiolum 
foliolis triplo breviorem equantibus, foliolis rigidulis, spathu- 
lato-linearibus, subtrigonis, cum mucrone rectis, 1-raro 2-jugis 
cum impari; floribus in apice ramulorum spicato-capitatis, 
breve pedicellatis, bracteolis 2 semi-connatis flori approxi- 
matis, calyce glabro subeequaliter 2-labiato. 
In solo argillaceo ad latus australe montis Tafelberg (III, 
A. e.) Sept. 1838. Hb. Krauss, propr. 
This seems to be related to P. triflora, Poir. DC. |. c. n. 6, 
and to P. affinis, Eckl. et Zeyh. en. p. 774, both which, how- 
ever, differ in having axillary flowers, and, besides, in the 
form and number of the folioles, etc. In our plant the leaves 
are 3-foliolate (except a very few of the lower ones which are 
pinnately 5-foliolate) and the leaflets are but 3-6 lines long, and 
$ line broad; they are acuminated into a straight sharp 
mucro, and their common petiole and stipules vary from one: 
to two lines in length. Flowers in very short capituliform 
terminal spikes (never axillary !) fine blue, quite of the form 
and size of those of P. verrucosa. Calyx brownish, like the 
leaves and branches densely punctate with glands, its lobes, 
especially of the upper lip, ciliated with short black hairs. 
To this we refer P. tenuifolia, Ecklon et Zeyh. enum. p. 225. 
(Un. itin. n. 658!) which differs from Dr. Krauss’s speci- 
mens only in having the top of the branches and the calyx 
slightly pubescent, the uppermost Jeaves sometimes unifolio- 
late, the folioles a little longer, and the flowers more remote 
and forming a very short terminal raceme. We must doubt 
whether it be really Linné’s P. tenuifolia, which is described 
as having “rami leves, foliola lineari-lanceolata” and “ pedi- 
celli axillares.”—P. filiformis Poir. ! (according to Poiret's au- 
tograph specimen in Lamarck's Herbarium, which we have 
compared with Ecklon's, and to which is ascribed as a sy- 
nonym * P, angustifolia Hort. Kew.") is, indeed, very like 
