Abstracts 59 



Eight subspecies of A. Sanguisorl>a< ;\xv created £91' the Australian and Tas- 

 manian forms, none of which extend to New Zealantl. The New Zealand forms, 

 all of which arc endemic, consist of the following subspecies :— 



1. Novae-Zelandiae (Kirk) Bitter, and its varieties— viridissima Bitt. and 



subtusglaucescens Bitt. 



2. Oaesiiglauca Bitter (probably = var. pilosa T. Kirk of .4. Sanguisorbae) 



and its vars. brevibracheata Bitt. and involucrata Bitt. 

 :?. Profundeincisa (described from cultivated plants) and its variety 

 sericeinitens Bitt. (Kelly's Hill. leg. L. Cockayne). 



4. Pusilla (described from a cultivated plant in the Bremen Botanical 



Garden) and its five varieties, three of which are founded on speci- 

 mens in Herb. Berol., collected by Krull in Chatham Islands, and 

 another var. antarctica Cockayne. 



5. Aucklandica* (Auckland Island, Hooker, f. Herb. Berol.. Florent,, 



Paris). 



The remainder of the New Zealand species arc put into two special sections ol 

 the genus, both of which are endemic. A. glabra Buchanan forming the section 

 pteracaena Bitter, and A. microphylla Hook f. and A. Buchanani Hook f. the 

 section microphyllae Bitter. A. microphylla is divided into the subspecies 

 eumicrophylla B'itt. and obscurascns Bitt. The former contains the var. inermis 

 (Hook, f.) Kirk, and this is resolved into two forms named respectively longiscapa 

 Bitt. and breviscapa Bitt. and the var. pallideolivacea Bitt. described from a culti- 

 vated plant in the Berlin Botanic Garden. The subspecies obsr.urascehs is based on 

 cultivated plants coming from the nursery of Thomas Ware ; the vars. depressa 

 T. Kirk and pauciglochtdiata Bitt. are included here. This latter is evidently the 

 dune form of Southland. A new variety inermis Bitt. of A. Buchanani is described. 



Finally, a number of hybrids of garden origin* mostly between New Zealand 

 species, especially A. microphylla and A. Sanguisorbae are described and their 

 leaves figured. 



It must be added that the author docs not look upon this work as a monograph 

 of the genus, but onlv as material for such. 



L. C. 



6. On the Peopling by Plants of the Subalpine River-bed of the 



Rakaia (Southern Alps of New Zealand). By L. Cockayne. 



(Trans, and Proo. Bot. Sue. Edin., vol. 24. pp. 104—125, pi. 3; 

 1911.) 



The relation between the evolution of a land form and its plant covering is a 

 matter of high phyto-geographical interest, but one extremely difficult to estimate 

 in the majority of cases. A New Zealand stony river-bed affords an excellent 

 subject for such a study. The peopling of such near its glacier source may be a 

 similar phenomenon to what happened on the Canterbury Plains at the close of 

 the glacial period. A brief account is given of the physiography of a river-bed, 

 and it is shown to be in a constant state of change, and to present all gradations 

 of station from new stony bed, swept at times by water, to low stable terraces. 

 The river-bed in question is in a forest climate depending upon the average dis- 

 tance reached by the western rainfall. Generally speaking, the climate is partly 

 hygrophytic and partly xerophytic, for the effect of the heavy rain is neutralized 

 by the insolation, the frost, and. above all, the high winds. An important factor 

 affecting the soil is the presence, at no great distance below the surface, of ice-cold 

 water. A glacial river-bed near its source is both a physically and physiologically 

 dry station. 



A synopsis is given of the species of the river-bed according to their growth- 

 forms. There are two low trees, fourteen shrubs, two lianes, and thirty herbs or 

 subshrubs. A special account is furnished of the Baoulia form, and it is pointed 

 out how the species show an epharmonic gradation of forms from the rapidly grow- 

 ing mats of B. tenuicaulis, with its open mesophytic leaves of seedlings and rever- 

 sion shoots, to the highly differentiated woolly masses of B. eximia, &c. Tin 



[* It is almost certain that this is identical with A. Sa"iui orlae var. nn'wdiia Cockayne. That 

 variety was founded, so far as the description of the flower and scape went, upon one flower coming out of 

 season on a plant just brought from Auckland Island. It is now known that the fruiting-scape is much 

 longer than as described, and that its hairiness was underestimated. Bitter suggests the Auckland Island 

 plant may be related to -4. msularis.] 



