COL. GRANT—BOTANY OF THE SPEKE AND GRANT EXPEDITION. . 87 
limbs; its powdered charcoal is rubbed into incisions made in the leg while sitting under the tree.— 
J. AG] 
18. Ixora (PAVETTA) ABYSSINICA, Fresenius, in Mus. Senck. ii. 166.— Pavetta, sp., 
160, App. Speke's Journ. 630. 
Var. Folia elliptica, breviter obtuse apiculata, utrinque subscabride hirtella ; calyx 
hirtus, lobis subulatis, tubo duplo longioribus; corolla tubo extus hirtello, lobis ob- 
longis obtusis 4—5-р1о longiore. 
Hab. Karagué, in flower, Dec. 1861, Col. Grant ! 
[Shrub, by water, Karagué. Inflorescence nearly a corymb. Flowers white, erect, platter-shape, 
* with long tube; calyx diminutive; the four stamina are exserted over the reflexed divisions of the 
corolla. Leaves leathery to the touch. In flower December.—J. A. G.] 
19. COFFEA ARABICA, L. 
[The native coffee was first brought to us at Karagué by the Wahia race, who market there; it is carried 
in bundles (made of plantain-leaf) the shape and size of an ordinary wine-bottle. We found it growing 
wild, and certainly never introduced, near the Victoria Nyanza lake at Witchroo, in May 1862, when it was 
in flower and scarlet berry. The bushes were 10 to 15 feet high, and afforded shade; the branches were 
arched over head from the weight of this and all previous year's fruit. Тһе roots arched above and out 
of the ground, extending even further than the branches. It grew on red clay soil, which was much washed 
away by constant rain in this equatorial region. Тһе inhabitants do not use the berry in decoction, like 
we do; they chew it, as an allayer of hunger and thirst—a soother; and I also found it have a pleasing 
effect if chewed. We saw no coffee equal in size to Ceylon Jackal-coffee; what was observed was an 
undeveloped bean with the husk on. I therefore infer that the bean is gathered long before it has come 
to maturity ; for it readily softens in the mouth. We found some huts by the side of the lake stored with 
quantities of it, but did not meet with it in any other portion of our route.—J. А. G.] 
20. SPERMACOCE DIBRACHIATA, Oliv., sp. nov. Herba 1-2-pedalis, erecta, scabrida; 
caule folioso, hispido ; foliis lineari-oblongis, acutis, supra glabratis, subtus in nervo medio 
hispidis; floribus in capitula multiflora, involucrata, pedunculata, terminalia dispositis ; 
bracteis duabus oppositis, basi in involucrum hemisphzericum hispidum connatis ; floribus 
tetrameris, in receptaculo brevissime pedicellatis, bracteolis linearibus paleaceis inter- 
mixtis; calycis limbo 4-partito, lobis elongatis, lineari-subulatis, rigidiusculis, scabridis ; 
corolle tubo gracili, lobis oblongo-lanceolatis, acutiusculis, apice setulosis; fructu 
oblongo-obovoideo, calycis limbo persistente coronato, lineis glandulosis, immersis 
notato. 
` Caulis erectus, ramosus, 1-2-pedalis, pilis subpatentibus subsparsis hispidus. Folia adscendentia, 
firmula, 2-3 poll. longa, 1—1 poll. lata, radicalia probabiliter latiora. Stipule vaginantes, dentibus seti- 
formibus. Pedunculi erecti, 3-4 poll. longi (in sp. Grant.)  Bracteæ involucri 14-2 poll. longæ, diva- 
ricate ; involucrum 3-1 poll. diam., hemisphæricum, hirsutum, receptaculo plano, paleis anguste linearibus 
bracteolato. Calyx lobis + poll. longis. Corolla pollicem longa, glabra, tubo gracillimo. Stamina 
— exserta, fauci glabri inserta. Stylus filiformis, stigmate capitato bilobo exserto. Fructus} poll. longus, 
lineam latus. ; | 
Над. Karagué, Feb. 1862 (Spermacoce no. 2, App. Speke's Journ. 636), Col. Grant ! 
A closely allied, if not specifically identical, plant occurs in the Manganya hills, Zam- 
besia, in which the peduncles (last internode under the involueral bracts) vary from ап 
inch or two to š foot in length. 
[Twelve inches high. Flowers (February) sky-blue; after they fall off, the sepals vio brown and 
| N 
