138 ECOLOGY, PLANT GEOGRAPHY [BoT. Absts., Vol. VIII, 



juncaceous Andesia hisexuali 15 to 30 cm. high forming a thick carpet. — Trees are absent 

 throughout, and even in the valleys the shrubs do not exceed 2 m. in height. Adesmia pini- 

 folia, a legume, is the most plentiful shrub, while among the others are Ephedra americana 

 andina, Berberis empetrifolia, and Senecio uspallatensis. Opuntia andicola, the only cactus 

 of the region, together with Azorella Gilliesii and Laretia acaulis, 2 umbellifers, form a 

 curious trio of herbaceous cushion plants confined to the valleys. — Upon the more exposed 

 parts of the mountains there is a notable abundance of prostrate, tufted, rosette and cushion 

 plants often with a striking development of large woody roots. These growth-forms are 

 accounted for as being in response to exposure to high winds and dependence upon a sub- 

 terranean water supply. Upon the slopes Adesmia trijuga, with shrubby cushions 30 cm. 

 high, together with Poa chilensis and Stipa speciosa in tufts, dominate the area, forming 

 scattered dots over the rocky landscape. Most abundant upon the summits between 3000 

 and 4000 m. are the subterranean woody cushions of Adesmia subterranea, whose leaves form 

 a carpet upon the surface. Accompanying this species with similar growth-forms are the 

 more uncommon Verbena uniflora and Oxalis bryoides. — The entire vascular flora consists of 

 417 species, including 1 pteriodophyte, Cystopteris fragilis, and one gymnosperm, Ephedra. 

 Among the best represented families are Compositae with 85 species, Leguminosae'with 36, 

 Gramineae with34, Cruciferae with28, Portulacaceae with 15, Umbelliferae with 15, Rosaceae 

 with 12, Cyperaceae with 12, Oxalidaceae with 10, and Violaceae and Caryophyllaceae with 

 9 species each. Well represented genera are Senecio with 26 species, Adesmia with 16, Calan- 

 drinia with 15, Astragalus with 12, Oxalis with 10, and Viola with 9. The scarcity of the 

 Saxifragaceae, with 2 rare species, and the entire absence of the Ericaceae and Primulaceae, 

 are worthy of note. Lichens, abundant at the lower altitudes, become very rare above 2800 

 m., mosses are common about springs up to 3600 m. but liverworts are entirely lacking. More 

 than one-half the species (210) are classed as belonging to the central Andes, GO being endemic. 

 There are no endemic genera but notable among this group are such aggregates as 6 species 

 of Adesmia, 2 of Boopis, 12 of Senecio, and 2 new varieties of Koeleria. The other elements 

 are the northern tropical with 16 species, the subtropical with 21 species, the basal Argentinan 

 with 56 species, the southern Andean with 10 species, the Patagonian with 73 species, and the 

 cosmopolitan and introduced species numbering respectively 28 and 17. This introduced 

 element must be regarded as small when it is recalled that the Mendoza River valley has 

 been the trans-Andean route for centuries. — There is a carefully and fully annotated list of 

 the vascular flora of the regions based on the collections of the author and those of other 

 explorers, and finally there is a bibliography of 54 titles. — Photographs and drawings illus- 

 trate many of the less known species. — Geo. D. Fuller. 



9.54. Rf BEL, Edtjard. Uber die Entwicklung der Geselischaftsmorphologie. [The evolu- 

 tion of the morphology of plant communities.] Jour. Ecol. 8: lS-40. 1920. The author gives 

 an outline of the history of statistical investigations of vegetation from Humboldt (1807) 

 to the present, tracing an evolution in the study of the forms of plant communities, and 

 treating the work of nearly 30 different workers in some 24 pages. — In different plant com- 

 munities quantitative results have been sought for such matters as abundance, dominance 

 and its sub-classes, manner of occurrence (gregarious, solitary, etc.), frequency, life-forms 

 (growth-forms) and layers, specificity of species to associations ("Gesellschaftstreue"). 

 The aim is the quantitative expression of relations obtaining in such communities for the 

 investigation of associations and succession and for the purposes of plant geography. The 

 varying forms of associations are well characterized by these results. There is no finality 

 in any method or combinations of methods. The future will undoubtedly show greater 

 advances. The author emphasizes the need for the study of forms in distinction from their 

 ecology. He states that hitherto the work has been too largely concerned with the ecology 

 of plant communities and that it should be more concerned with the morphology of plant 

 communities in order that the results obtained may be applied in the elucidation of succes- 

 sions. — The paper brings into view, in its sketch of the evolution of statistical methods for 

 over a century of activity, that the quantitative results so far reached are sometimes really 

 quantitative only in the sense of being numerical description.— i/. de Forest. 



