50 REVIEWS. 



For purposes of comparison the world is divided into fifty " regions," 

 each supposed to indicate a more or less marked flora, and defined 

 by strougly marked geographical limits and climatal peculiarities. Thus, 

 there is a well-marked arctic region comprised within the polar circle, 

 and common to Europe, Asia, and America. Many of the species 

 characteristic of this region appear again on the limits of perpetual snow in 

 more southern climates, and yet the alpine flora, taken as a whole, is very 

 distinct from the arctic. Europe, minus its arctic portion, is divided into 

 two well marked regions, the temperate or Northern and Midland States, 

 and the Mediterranean, including the shores of that sea, as well as the 

 whole Iberian peninsula. Some of the detached Atlantic Islands, such as 

 St. Helena (with Ascension), and the minute Tristan D'Acunha, are distin- 

 guished as separate regions, and apparently with justice, so very peculiar is 

 their flora, especially of the former. Such minute specific centres contrast 

 strongly with our author's 29th region, which comprises the Australian 

 continent, with its outliers, Tasmania, New Zealand, Norfolk Island, and 

 New Caledonia. North America, east of the Rocky Mountains, and as 

 far south as the mouths of the Mississippi, forms another extensive, but 

 well marked region. Brazil is divided into three regions, and the remain- 

 ing portion of S. America into ten ; so varied is that continent, in different 

 parallels, in its vegetable productions. The author does not lay much 

 stress on the exactitude with which his regions represent natural limits ; 

 indeed, he states that his arrangement is defective on several points, but 

 having commenced to tabulate on this basis it was necessary to go on with 

 it. Tables are given illustrating the area occupied by species and families. 

 From these it appears that the comparatively small family of Papaveracew 

 includes some of the most widely dispersed plants, one of its species (Arge- 

 mone mexicana) occurring in eleven regions, while Myrtaceae, a very ex- 

 tensive order, is specifically very local, only three out of its 700 species 

 occurring in three regions. Of Papaveraceae, 68 per cent., Cruciferae, 75 

 per cent., Campanulaceae, 89 per cent., and Myrtaceae, 97 per cent, consist 

 of species limited to a single region. Next follow tables contrasting the 

 floras of different countries, the first table, illustrating that of New Holland, 

 being copied from Mr. Brown, who states, that out of 3,760 flowering 

 plants known (in 1814) as natives of Australia, 45 are common to Europe. 

 Since that date the number of Australian species has been doubled, but not 

 many more European species have been detected. Perhaps one of the most 

 remarkable discoveries of late years has been that of Lysimachia vulgaris, 

 found by Dr. F. Mueller in the alpine region of Victoria, far removed from 



