163 



species somewhat smaller than the number of summer-annuals. 

 In the case of several of the other countries given in the 

 table, there was the same difficulty in obtaining reliable 

 comparative data as to the life-period of the plants, therefore 

 I am obliged to limit the table to a comparative statement of 

 the flowering season for all the countries. This table will 

 to some extent illustrate the proportion between ephemerals 

 and summer-annual plants. 



The four upper lines are given for comparative purposes. 

 The figures for Denmark are calculated from RAUNKLER'S 

 Excursion-Flora, for Yekaterinoslaw according to BEKETOFF, 

 for Samos according to STEFANI, FORSYTH MAJOR & BARBEY, and 

 HALACSY, and the figures for Spain according to WILLKOMM 

 (1852) 1 ). No figures are given for the Libyan desert, as the 

 flowering season is not recorded in ASCHERSON and SCHWEIN- 

 FURTH'S flora, also because many of the annual plants are 

 facultative biennials or perennials, and some flower both in 

 spring and winter. The following species of annual summer- 

 plants comparable to those occurring in the Transcaspian 

 desert i. e. not facultative perennials are given by 



VOLKENS for the Libyan desert: Moricandia clauata, Diplotaxis 

 Harra, Monsonia nivea, Mesembryanthemum crystallinum, M. 

 nodiflorum, M. Forskalei, Aizoon canariense. Although this 

 list be not complete, it indicates that there are very few 

 summer-annual plants in the Libyan desert, so that probably 

 this locality should have been placed at the foot of the 

 table, being the one with the greatest number of ephemeral 

 plants. 



The desert of western North America might be placed 

 in the table beside Transcaspia. According to THORNBER 

 there are in the desert of Tucson about 20 per ct. "long- 

 lived annuals" and 80 per ct. short-lived ones. By counting 

 up SPALDING'S list (1909) of the plants of the Tucson country 

 I obtained respectively 21 and 79 per ct. of these two cate- 

 gories. The proportion thus agrees approximately with that 

 found in Transcaspia. 



') See also WILLKOMM 1896 p. 150. This work cannot be employed 

 for statistics, as all the species of the formations are not recorded. 



11* 



