728 



SCIENCE 



[N. S. Vol. XLI. No. 10C3 



important to note that not one has been able to 

 establish itself under constant natural con- 

 ditions. Of the remaining 131, all but sis or 

 seven are more or less widely distributed, chiefly 

 in the surrounding country. But these few 

 appear to be endemic, as they have not been 

 found elsewhere. The suggestion is obvious 

 that these have originated in the sink during 

 comparatively recent times, while it is further 

 pointed out by Dr. MaeDougal that other 

 species may have similarly arisen, but have suc- 

 ceeded in passing outwardly beyond the limits 

 of their original home. There is an approach 

 here to something like quantitative relations 

 between geological age and the possible num- 

 ber of new specific origins. 



It seems equally probable that other plants, 

 such as the desert palm Washingtonia filifera 

 and Populus Macdougalii, are to be referred, 

 as to their origin, to comparatively recent 

 dates, and this locality. 



The absence of succulent serophytes, in- 

 cluding under this term those with water- 

 storage roots, from this very pronounced desert 

 region is worthy of remark, since, in the minds 

 of m.any, succulence is regarded as the final 

 expression of desert adaptation. Here the 

 xerophytic shrub with spinose parts and other 

 appropriate characters are the chief perennial 

 inhabitants of the slopes and older strands, 

 while the salt-laden alluvium of the sink-floor 

 bears a zone of the salt-bushes, Atriplex spp. 



The flnal paper of the series concerna the 

 movements of the vegetation due to submer- 

 sion and desiccation and is by Dr. D. T. Mac- 

 Dougal, under whose leadership the whole 

 work has been carried forward. Recognizing 

 the importance of the opportunity to observe 

 the advance of plants upon an immense 

 sterilized area especially in view of the inade- 

 quate study or total neglect of analogous 

 earlier opportunities (one thinks of the lost 

 one of Mont Pelee), the lavas of Hawaii, 

 studied by 0. N. Forbes excepted, the task 

 was laid out on a comprehensive but workable 

 scale. Sample areas or "belt transects," a 

 mile in width, normal to the beach lines, were 

 chosen, and these, together with sterilized is- 

 lands, afforded the basis for exhaustive study. 

 This, as the reader will have understood from 



what has already been said, embraced not only 

 the vegetation, but the salt content of soil and 

 water and other relations. Usually semi- 

 annual visits were made for the collection, of 

 data. 



The flrst half of the paper presents the 

 facts concerning the reoccupation of the 

 strands of six successive years, and a partial 

 study of another, namely, 1913. The earlier 

 strands of Blake Sea, untouched by the recent 

 invasion of waters, afforded a standard for 

 comparison, so that it was possible to measure 

 the rate at which the f acies of the new strands 

 took on the same composition as obtains now in 

 the old, relatively static strands. It was ob- 

 served that the recession of the water was so 

 soon followed by desiccation of the soil that 

 wholly desert conditions were established in 

 the course of a couple of years, and that, in 

 consequence, the introduction of xerophytes 

 identical with those characteristic of the 

 ancient Blake Sea strands had been accom- 

 plished in the course of three or four years. 

 The change from close to open formation was 

 similarly rapid. 



The transition from one environment to an- 

 other as the established desert gives way to 

 strand, and the gradual alteration of successive 

 zones correlated with the recession of the 

 water, together with the separation of shore 

 and sterilized islands by extensive water ways, 

 sets up conditions for the study of methods of 

 dissemination and of natural selection as well 

 as reoccupation. It is of more than incidental 

 importance that the reoccupation of islands, 

 and of one shore from another, was among 

 other methods possible chiefly by the flotation 

 of seeds and fruits as proved by many experi- 

 mental tests. It is clear that in this can be 

 seen no causal relation between the conditions 

 and the " adaptations to flotation." Nature 

 had otherwise been peculiarly far-sighted in 

 furnishing to desert plants not only adaptations 

 in harmony with their immediate surround- 

 ings, but with a possibility so remote as the 

 occurrence of a lake! Causal relations are, 

 however, to be seen probably in such characters 

 as reduced superficies, thickened outer tissues, 

 and the like, as a direct result of evaporation, 

 and a number of such correlations have been 



