Morphological Observations. -97 



second essential characteristic, the comparative late appearance, was 

 also strikingly shown. While there were 10 primordia of leaves clear- 

 ly shown in one preparation, only in the first five the intrafoliar 

 structures were developed, the remaining 5 were still without them. 

 There is, therefore, no reason why we should not consider them 

 as stipules in the cases where they are leaflike in character. 

 I have .satisfied myself also that in A. ■filament osa where they appear, 

 as in other species belonging to section Telcphiastrum. as hairlike 

 structures, they take their origin from the base of the primordia of 

 the leaves as comparatively thick protuberances, on which the hair- 

 like structures which eventually only are visible to the naked eye 

 arise at an early stage, though again only some time after the pri- 

 mordium of the leaf has become clearly differentiated. 



In his well-known work. " Die Flora der aegyptisch-arabischen 

 Wiiste "' (Berlin 1887) Georg Volkens says: "A considerable number 

 of desert-plants exhibit in the mesophyll of the leaves, sometimes also 

 in the intemodes (Tamarix), water-storing elements which were dis- 

 Though they are common in desert-plants, it appears from Volkens' 

 called them resen-oirs vasiformes." Heinricher devoted a special paper 

 to them and called them " Speichertracheiden." which I will trans- 

 late " Water-storing tracheides." Unfortunately I have had no access 

 to Heinricher's paper (Bot. Centralblatt. 1885, XXIII, No. 27/28). 

 Though the} are common in desert-plants, it appears from Volkens 

 work that nowhere are thev so well developed as I found them to be in 

 Anacampseros. at all events in the shoots of A. Alstomi. A. -fHatnen- 

 iosa and A. arachnoides. They are absent in the leaves and the 

 cortex of these shoots. The vascular bundles are very poorly develop- 

 ed in the branches of the species which I have examined. Apart 

 from them. j)ractically the whole of the central cylinder is composed 

 of tracheides in A. Alstonii. In A. papyracea they are also ven 

 numerous, but do not extend to the centre which is composed of 

 parenchymatous tissue, thus forming a hollow cylinder. In A. filamen- 

 tosa they do extend to the pith, but are freely interspersed with 

 parenchymatous cells. Their pits are usuall\' scaJariform. Even if 

 we had no previous investigations on the subject, we would be led 

 to the conclusion that they serve the purpose of storing water, as 

 their development is such that their presence for merely strengthen- 

 ing the stem or for conducting water would be unintelligible. 



There are other differences in detail when we consider the 

 anatomical structure of the stem, but there is only one which need 

 concern us here. In A. Alstonii and A. papyracea. in which the 

 assimilatory system is so- very much reduced and where it is further 

 so well protected by the stipules from excessive transpiration, we find 

 that the shoots are not protected by cork while in the two species 

 belonging to the section Telephiasirum to which I have just referred 

 there is a well-developed and very active cork-cambium which soon 

 provides them with a strong peripheral layer of cork. 



In the section Avonia the caudex nO' doubt also serves as an im- 

 portant water-reservoir, but I have not examined its anatomical 

 structure yet. 



