CH. xvii] AND GENERA OF LARGER SIZE 191 



but it must suffice to have drawn attention to them, and to the 

 very clear way in which they show that on the large scale and in 

 the long run distribution is a very mechanical process, i.e. that 

 the various factors causing it act at a very imiform rate, and 

 that it is usually only stopped by actual barriers. 



The individual area occupied by a monotype genus may vary 

 enormously, but is usually rather limited. 1037 of them occur 

 upon islands, and when the island is of any large size are usually 

 restricted to a portion of it. The great bulk of those mentioned 

 as found only in South or North America, or in x^frica, and the 

 241 of Australia, are similarly restricted, and so are most of 

 those in the other great regions of the globe. When one comes 

 to genera found in both worlds, one finds that only 66 of them 

 are monotypic, or a mere 5-6 per cent, of the genera that occur 

 in both. All but about 20 of these are found only in the north 

 temperate zone, which by reason of its connections by way of 

 the arctic regions, formerly passable for plants, is not really so 

 large in proportion as it seems. Bolhoschoenus, Brasenia, Hakon- 

 cchloa, Hijypuris, Montia, and Zannichellia are more or less 

 cosmopolitan, and of the remaining genera three are coastal 

 plants carried by sea currents, and four are tropical American 

 and West African— countries united by a current that crosses in 

 about three months. Pistia is a water plant, and some of the 

 others are doubtful identifications, so that there remain a bare 

 half-dozen that have a very large range, evidently acquired by 

 land, or nuich less than 0-25 per cent, of the total of monotypes. 

 These are Christiana, Eidophidium, Manisuris, Remirea, Rliabdia, 

 and Sphenoclea. In fact, it is fairly evident that if one were to 

 determine accurately the areas of the 4853 monotypes, one would 

 obtain the usual hollow curve, beginning with a great many of 

 very small area, and tapering aAvay to the other end as areas were 

 reached of larger and larger size. 



In any country in which there are many monotypes, their 

 areas tend to overlap like those of the endemic species. Thus in 

 New Zealand, in any zone of 100 miles from north to south on 

 the main islands, there are never less than se\^en monotypic 

 endemic genera, though of the eighteen such genera six are 

 northern, ceasing towards the south, and twelve are southern, 

 two only of which reach the far north. Just as with the species, 

 the genera show a maximum number about the centre of New- 

 Zealand. What reason (in adaptation or relic nature) can one 

 find for the fact that one genus reaches from the far north to 



