REGIONS AND AREAS OF DISTRIBUTION. 559 
whole nineteen or twenty Composite this will probably prove to 
be the only one presenting a remnant of an ancient flora, the only 
exception to the more or less adventitious character of the 
remainder. 
5. Atlantic Islands. 
Under this head are included the Azores, Madeira, the Canary 
Islands, and the Cape-Verd Islands. Their flora has been well 
investigated ; and there are probably but very few species that 
have escaped notice, although there is much to be learnt yet as to 
the limits of variation which the native species have undergone, 
and different botanists have from the same materials very variously 
estimated the numbers of the principal genera, such as Chrys- 
anthemum, Senecio, Sonchus, Tolpis, die. The following Tables 
are deduced from data supplied by Seubert’s ‘Flora Azorica,’ 
Webb’s * Phytographia Canariensis’ worked up as to Composite 
by Schultz Bipontinus, Lowe's * Flora of Madeira, and Webb’s 
* Spicilegia Gorgonea’ inserted in Hooker's * Niger Flora,’ with a 
few modifications, chiefly reductions of species, suggested by the 
specimens preserved in the Kew herbaria. As a general result, 
these islands, in respect of Composite, may be considered as an 
outlying district of the rich Mediterranean region, with a very 
slight tendency in a few Canary forms to a South-African type, 
without, however, any identical or even representative species, 
and in the most southern, the Cape-Verd group, an admixture of 
a few genera or species belonging to the neighbouring tropical 
African continent. More of the introduced species are here 
admitted than in the Tables of Continental distribution; for in 
islands introduced plants sometimes form so essential a part in the 
vegetation that they are inserted without remark in the local 
floras, and I have not been always able to distinguish them from 
more ancient inhabitants. I have, however, omitted such evident 
escapes from gardens as Tagetes and Zinnia. 
LINN. JOURN.—BOTANY, VOL, XIII. 
