8 
James Small. 
account of the family, so that only a general outline of the more 
striking features will be given. 
A map (Plate I, Fig. 29) is given in which the areas of the more 
important species are delimited by a thick red line, those of less im¬ 
portant species with more than a local habitat by a thin red line and 
those of species which are limited to one state or country by red 
dots. Accuracy in detail has not been aimed at on account of the 
small size of the map and because it is unnecessary for the 
elucidation of the chief points, namely the centre of origin, the 
paths of migration and the centres of concentration. 
The Centre of Origin. 
In a number of other widespread genera the centre of origin 
is clearly indicated by the coincidence of the point of overlapping of 
the areas of widespread species with the centre of concentration 
(see PI. II, 31, PI. Ill, 34, Pl.V, 37,38), but in Senecio there is no such 
indication. Evidence will be given in Chapter XI for the origin of 
StfMecfoinoraboutthe Bolivian region. The north of South America is 
the probable centre of origin for most large Angiospermous families 
(cp. 4). One point which confirms this hypothesis for Senecio is the 
very large concentration of species along the Andes and on the 
campos, savannahs and the margins of the Brazilian forests, hut as 
there are other equally intense concentrations in South Africa and 
Mexico this does not form conclusive evidence. 
Taking this point as granted in the meantime we can consider 
the South American species. The species with only a local dis¬ 
tribution number in Chili 318, in Peru 66, in Bolivia 56, in Ecuador 
36, in Colombia 45, in Brazil 56, in Argentina 36, and in Patagonia 
46. These numbers are, however, of little importance for com¬ 
parison among themselves as they depend perhaps more on our 
present knowledge of each country than on the actual number 
of species present. The large number in Chili, for instance, is due 
to the activity of two botanists Remy (130-131) and Philippi (126- 
127). It is clear, however, that the Andean is a particularly 
favourable region for specific differentiation in the genus. On the 
other hand, not a single species of Senecio is recorded from 
Honduras, our knowledge of the flora of that Central American 
state having been limited until recent years to a list of some 
seventy species mentioned by Morris (54). More recent collections 
have added somewhat to our knowledge of this flora (cp. 41 & 83). 
