rK(\M AX OLD STANDPOINT. 



177 



tliose cases where the faiiiihes coiicenied, though each in its 

 own region, hold much of the globe between them. A striking 

 instance of this has just been given: and another is noted 

 below ill connection with the primitive group of the Scita- 

 mineiv. 



A good test of the erticacy of the differentiation theory is 

 aftbrded by those families that are widely spread over the 

 warm regions of the earth, yet stand well apart from all other 

 faiuilies. Such families occupy regions now separated by the 

 breadths (jf the Pacific and Atlantic oceans. Let us, for 

 instance, take the I'almaceie. The palms are numerous in 

 tropical Asia and in tropical America, and we know that they 

 extended in mass much farther north during the Tertiary 

 period. What encouragement, therefore, can the facts of 

 distribution afford us in searching for the home of the family, 

 when they indicate that the farther we go back the wider is 

 the range ? To attempt it would be at once to involve oneself 

 ill a labyrinth of assumptions both geographical and botanical. 

 Far fewer ditticulties would attach themselves to the explana- 

 tion supplied by the differentiation hypothesis that there 

 was originally a world-ranging palm prototype which lias 

 differentiated i/i ^Ita in various regions, and that its present 

 concentration in equatorial regions is connected with the 

 ditterentiatiun of the clhnate of the globe. 



The Palmace:e offer a suitable and familiar illustration of 

 the argument here followed ; but it would be easy to mention 

 other tropical families distributed around the globe, wliere the 

 attempt to discover a centre of development would be equally 

 futile.* This could only be in any degree successful in the 

 case of those more localised tro})ical families which belong to a 

 group of closely related fainilies, and are really the tribes of a 

 primitive family that has disappeared in the process of 

 differentiation. A good example is afforded in the case of 



^ As another mstaiice I will take the Monimiaceje, a family confined 

 to the tropical and subtropical regions of the Old and the New World, 

 and described by its recent monographers (Perkins and Gilg in Das 

 Pjianzenreich^ iv, 101, 1901) as so well defined and so natural in its 

 characters tliat all its species may be regarded as derived fiom a 

 single Old World stock that probably had its'birthplace in Indo-Malaya. 

 Since, however, all the fiv^e tribes are common to the Old and the New 

 World, whilst (jne-fourth of the genera and two-thirds of the species are 

 purely American, such an explanation raises a host of difficulties. 

 According to the ditterentiation hypothesis, we should merely begin with 

 a primitive parent type originally \l iff used in both the Old and the Xew 

 World and subsequently dilierentiating at unequal rates. 



