122 



The Journal of Heredity 



shell-like endocarj) around the kernel, 

 as may have been the case with the 

 fossil seeds described by Dr. Berry, to 

 judge from the fcjllowing statement in 

 the description of the fossil : 



"The surface is longitudinally wrinkled, due 

 possibly to desiccation before preservation, 

 which may also make the dimensions as given 

 probably under what they were in life. The 

 flesh was relatively thin compared with that 

 of the cultivated date and must have been of 

 considerable consistency and fibrous rather 

 than of the soft and almost fluid character of 

 some of the modern varieties of the latter." 



The very wide distribution of the 

 species of Phoenix in the Eastern 

 Hemisphere makes it the more difficult 

 to credit a complete extinction of the 

 groujj in the Western Hemisphere. 

 Date palms arc known from all of the 

 tropical and subtro])ical regions of the 

 Old World, from India through Africa 

 to the Canary Islands, though they do 

 not extend far into the Chinese region, 

 to Australia, or to the islands of the 

 Pacific. This is in contrast with the 

 distribution of the Asiatic fan-palms, 

 which are well represented in eastern 

 Asia and the Pacific islands. With the 

 possible exception of Raphia, no other 

 genus of palms is so widely distributed 

 as Phoenix, either in the (Jld World, or 

 in the New. The only comjjetitors in 

 this respect arc the palms that may 

 have had htiman assistance, such as 

 Cocos, Acrocomia, Attalea and Inodes. 

 The last is the most ubiquitous in the 

 wild state in the American tro])ics, and 

 in this respect also may be considered as 

 the American analogue of Phoenix. 

 These two genera enjoy to a greater 

 extent than most of their relatives the 

 same adaptive advantage, that the 

 seeds retain their vitality for long 

 periods in the dry state. With most 

 palms drying the seed soon kills the 

 enbryo. 



The relations of Phoenix with the 

 Old World fan-palms might also have 



bearing u]3on the jilace of origin. 

 Students of palms have generally agreed 

 that Phoenix is most closely related 

 with Chamaerops, a genus confined to 

 the Mediterranean region, the Asiatic 

 species formerly included in Chamaerops 

 being assigned by recent authors to 

 Trachycarpus. Chamaerops has larger 

 and more date-like fruits than Trachy- 

 carpus, and stronger sj^ines on the 

 petioles. The production of hybrids 

 between the date palm and Chamaerops 

 has been claimed, but is difficult to 

 credit in view of the ntunerous struc- 

 tural differences between the two palms. 

 The date palm differs from all of the 

 related fan-palms in having the sexes 

 on separate plants, and in ha\4ng the 

 inflorescence highly specialized. The 

 branches are mostly simple, and are 

 arranged in clusters or whorls, a con- 

 dition that may have been attained by 

 uniting the basal joints of the ])rimary 

 branches with the main axis. Another 

 profound difference is the reduction of 

 the spathes to a single one, which en- 

 closes the entire inflorescence until the 

 time of flowering. 



If the fossil seeds from Texas represent 

 a true date ]3alm, it may be expected 

 that remains of other Asiatic genera of 

 fan-palms will be found in America. 

 On the other hand, if the fossils repre- 

 sent a collateral relative of the date 

 palm the separation between Phoenix 

 and the Old World fan-jialms may 

 ha\'e occurred in the Eastern Hemi- 

 sphere. In view of the analogy pre- 

 sented by Brahea, the fonner ]3resence 

 of Phoenix in America can hardly be 

 detennined from the seeds alone. Other 

 parts of the plant are needed to make 

 the identification sectire. It is to be 

 hoped that nothing in the way of palm 

 materials will be o\'erlooked by those 

 who have the opporttmity to collect 

 fossil i)lants in the Southwestern 

 States. 



