329 R. A, Dümmer. 
One sheet bearing a specimen of A. imbricata Willd., and A. villosa 
Willd. (form), the former labelled in Linne’s handwriting (ink) , Har- 
togia imbricata“, the latter pencilled off and in Smith’s pencil, ,est D. 
ericoides“. 
A sheet with a form of A. imbricata Willd. labelled above in large 
letters ,Barreria“ below „capensis“, both in Linne’s handwriting. This 
is Barreria capensis Linn. Sp. Pl., 274 (1753). 
One sheet bearing a specimen of A. imbricata v. acuminata Sond., 
below an appended note in Linne's handwriting, „Diosma No. 6? Species 
— ciliatae varietas — Hartogia^; (the preceding three sheets are pinned 
together). 
One sheet bearing three flowerless twigs peraphs belonging to 
forms of A. imbricata Willd. „In Linne's handwriting below —“ Sparm. 1’ 
29 — capensis; what the numbers after Sparmann's name denote is 
unintelligible to Dr. D. Jackson. 
One sheet of a fine specimen of A. rugosa Link. bearing below, ‚sp.‘ 
and ,lanceolato* in Linne's handwriting. 
One sheet of a small specimen of A. Wrightii Macowan, bearing 
below, Spar.!) 35 and ,Hartogia lanceolata^ in Linne's handwriting. 
One sheet of A. craspedota E. Meyer (a fine specimen) not named 
and bearing the number 114 on an appended scrap of paper below. 
Two sheets of Agathosma ciliata Link, each bearing one specimen, 
one a probably cultivated specimen from Cliffort’s garden, with a large 
lettered „Diosma“ above, a small-lettered „ciliata, 5“, below, both in 
Linne's handwriting; the other only labelled „ciliata“, also in the author's 
handwriting. 
One sheet of a small specimen of A. erecta Baril. and Wendl., la- 
belled „capensis“ underneath, with a latin diagnosis on the reverse side. 
One sheet of A. orbicularis Bartl. and Wend. not labelled. 
1. Agathosma tabularis (§ Barosmoides) Sonder in Harv. and Sond. 
Fl, Cap., I, 403 (1860) = Barosma pulchella Eckl. and Zeyher var. tabularis 
Diimmer, nov. comb. 
Both Sonder, and Bentham and Hooker allude to the clawed or 
clawless petals as points of generic importance separating the genera 
Barosma and Agathosma, but these characters are prevalent in both ge- 
nera, as instance the § Barosmopetalae of the latter genus, where claw- 
less petals obtain, while in the preceding and subsequent sections of 
the same genus the reserve is the case; similarly the equality or in- 
equality of the calyx-lobes or calyx-segments are characters unworthy of 
generic significance, The most reliable characters separating these closely 
allied genera appears to me, (after a careful examination of both groups) 
to be the disposition of the flowers. In Barosma they are axillary and form 
collectively a racemose inflorescence, while in Agathosma they are more 
or less terminal and constitute a capituliform or umbellate inflorescence. 
1) Spar. denotes Sparmann the collector, a former pupil of Linneus. 
