168 A. D. Hardy: 



length — from a tenth to a twentieth part. In many twigs the 

 spiral growth was at or near the base, in some others about midway, 

 and in a few the terminal node was the abnormal one. In Casua- 

 rina the staminiferous twigs have the stamens at the nodes of the 

 apical end only, but in the abnormal twigs a spiral staminiferous 

 node was in one case succeeded by several nodes of purely vegeta- 

 tive character. Staminiferous spirals were usually terminal, and 

 the anthers and their pollen grains were morphologically good; so, 

 too, were the stamens borne at terminal nodes of twigs affected by 

 torsion nearer the base. The method of growth of these spirals 

 appears to be as follows : — Instead of the usual production of 

 Avhorls of leaves, which in the matured branchlets might reach an 

 inch or more in length — laterally connate and decurrent except 

 for the scale-like free end which forms the cup whence springs the 

 succeeding shoot — there is, usually, in the abnormal branchlets a 

 bursting of the cup-like circlet of scales, and an oblique emergence 

 of a laterally-developing spiral band, forming a t-ortuous structure 

 with laterally connate members, each of which is terminated by a 

 pointed, scale-like leaf-end, similar to those of a normal whorl. 

 The leaves laterally connate in such a laterally winding spiral are 

 one-fourth the length of normal branchlets, and in number may be 

 regarded as indefinite, there being 45, 51, and 59 respectively in 

 three of the longer spirals which I closely examined; and other 

 spirals were longer. The spirals wound indifferently to the right or 

 to the left in respective tw^igs. The stamens in the case of a 

 staminiferous spiral appeared as a continuous fringe at the over- 

 lapping edge of tlie imbricated tunic so formed. The numlDer of 

 leaf-ends in a whorl in C. sfricta is not constant, but is from 9 t-o 

 12. - 



In many aquatic or marsh plants torsion of vegetative shoots 

 which are normally cylindrical and hollow is a not uncommon occur- 

 rence. H eleocharis sphacdata is one that I have frequently noticed, 

 and in this case the discoid septae become ellipsoid. The cylindrical 

 shoot becomes flattened, and twists in a more or less easy torsion 

 while keeping perfectly straight (the twist being that of an auger 

 rather than that of a corkscrew), while, in other plants, solid, 

 angular shoots may twist through an inch or two of the apical 

 end, the twist being a compromise l>etween a zig-zag (in one plane) 

 and a corkscrew spiral. This is exemplified by the shoots of 

 Xanthorrhoea minor (Plate XIII., fig. 5), the leaves, straight for 

 about 12 inches, having the terminal inch torsive. 



