300 Report S.A.A. Advancement of Science. 



attraction of insect-fertilisers. As far as my observations go, it is- 

 (jnly utilised by the plants to press the anthers against the stigma 

 when the flower closes again after having taken a short airing. It is., 

 however, very desirable that accurate observations on this point 

 should be made in their native habitats. 



People who are fond of theorising will naturally conclude from 

 my observations that Anacampseros is a genus which is passing away 

 from the state in which it required insects for fertilisation, as the 

 show-apparatus must at one time have been of considerable use in 

 attracting insects. Apart from the want of observations on this 

 point in the field, serious objections might be raised to this view, 

 but in one species at all events, in A. papyracea, the show-apparatus- 

 does not open out at all any more. We read already in the Flora 

 Capensis (II.. p. 383) that in this .species the flowers are included; 

 in the uppermost .stipules, but this does not express very clearly 

 what really happens. The flowers are not only included in the- 

 uppermost stipules, but they are closely covered over by them ; 

 they never open, and as they pretty regularl\ produce ripe ca])sule.s 

 and seeds, they are strictly cleistogamoiis. I have watched this 

 sj^ecies for years, and have never seen any flowers behave differentlv. 

 A. papyracea is, therefore, as far as we know, the only plant in 

 which none but deist ogamous flowers are produced. 



Fruit and Seed. — Some time after fertilisation, and shortly 

 before the seeds are ripe, the capsule is raised by the elongation of 

 the pedicel. This is especially striking in A. papyracea, as in this 

 way the capsule becomes free from the uppermost stipules. The 

 •structure of the capsule is pretty uniform in all species, differing only 

 111 unimportant details, but curiously it has. as a rule, been incorrectly 

 described. Sonder in the Flora Capensis (II.. p. 382) describes the 

 cap.sule as " conical, i -celled. 3-valved. the valves often longitudin- 

 ally divided, and then apparently 6-valved." 



Pax in Engler's Xatiirliche Pflanzenfamilien (111., ib, p. 57) 

 sa) s : — " Capsule conical or oblong, 3-valved. with fleshy epicari> 

 and membranaceous endocarp, valves often longitudinally divided, 

 and then apparently 6-valved." Bentham and Hooker (I.e. I., p. 157) 

 give a fairly accurate description, in which they follow Fenzl. 



In all species the two sepals and the petals with the attached 

 stamens close firmly together into a more or less cylindrical structure, 

 which eventually becomes detached at its base in the form of a cap. 

 Inside this structure the capsule is developed. The pericarp divides, 

 into an outer fairly hard, almost horny {not fleshy) epicari>, which also 

 becomes detached at its base in the form of a conical cap, and fre- 

 quently remains attached to the conical cap formed by the floral 

 envelope. This cap, formed by the epicarp, is split at its base 

 sometimes in three parts, in other cases in six parts. In the case 

 of A. Alstonii. each of these six parts is again slightly divided at its 

 base. There remains now. when the seeds are ripe, a basket-like 

 .structure, open above, formed, in the case of A. Alstonii, bv six 

 connivent oblong lobes, chiefly composed of strong fibres, and the 

 interval between them occupied by a single fibre, the whole- 



