598 Prain.—A Review of the Genera 
larger than in any Erythrococca and densely velvety outside, are again 
green and coriaceous when ripe, and open loculicidally without shedding 
their valves to expose the seeds, which in this case are bright yellow. In 
the case of the section Enclaoxylon (including Gynmoclaoxyloii) the fruits 
and seeds are, at least as a rule, indistinguishable from those of Erythro - 
cocca, though it is stated by Hooker (Flor. Brit. Ind., v. 410) that in Clcioxyloii 
there are species with cocci which, besides being coriaceous, may be indehiscent, 
and that in some of the species the seed may have no arillus. 1 It is true 
that ultimately in Claoxylon , as in Erythrococca , the two half valves of the 
cocci fall away, but there is no splitting of the septa such as is met with in 
the species of the allied genus Micrococca , to the fruits of which the descrip¬ 
tion given by Bentham is strictly applicable. 
A character, more interesting than important, which at first sight 
appears almost distinctive of E. aculeata , on which the genus Erythrococca 
was based, deserves to be noted. The pedicels in the male flowers of 
Erythrococca are articulate. The articulation is usually opposite or below 
the tip of the corresponding bracteole ; the portion of the pedicel under the 
joint is, as a rule, appreciably thicker than the portion above; 2 in cases 
where the rachis is pubescent while the pedicel is glabrous, the thicker 
pedicel base is often pubescent like the rachis, and not glabrous like the 
rest of the pedicel. In E. aculeata the joint is considerably above the level 
of the bracteole. But this feature is not peculiar to E. aculeata ; it is well 
marked in Claoxylon Menyharthii , which differs considerably from E. acu¬ 
leata , and in E. natalensis and E. berberidea , which differ considerably from 
both the foregoing. The meaning of this difference in the position of the 
articulation from that which usually obtains is not obvious; it is, however, 
so far as classification is concerned, only of specific value. 
Another curious character met with in E . acideata is that the male 
calyx is partite only to the middle, leaving below the teeth a distinct cam- 
panulate tube. In two other species, E. berberidea and E. subspicata , the 
calyx is two-thirds partite, but in the genus as a whole the calyx is partite 
almost to the base. This departure from the general rule is associated 
with a divergence from the usual conditions as regards the receptacular 
glands. In E. acideata these are unusually large, as long as the stamens; 
in E. berberidea and E. subspicata they are also large, as long as the fila¬ 
ments ; as a rule these glands are relatively small, generally minute. The 
horizontal plane to which division of the calyx extends in each case prac¬ 
tically coincides with the level of the points of origin of the viscid apical 
1 The possible association of exarillate seeds and indehiscent cocci is one that will doubtless re¬ 
ceive the attention it merits from whoever may prepare the revision of which the genus Claoxylon 
stands urgently in need. 
2 I11 the case of C. columnare , Muell. arg., the condition has been fully and correctly described 
by Mueller (DC. Prodr., xv. 2, 776); the character, however, is a generic and not, as Mueller’s 
treatment seems to imply, a specific one. 
