Ill 



of Chamaephytes (141), and Bews '^'^ in the Flora of Natal notes the same 

 thing (19 per cent.). Raunkaier has shown that Chamaephytes increase towards 

 the polar regions and also in hot climates with a dry season. Even allowing for 

 the fact that the Pearson Islands have a restricted flora and a special type of 

 habitat, the relatively large number of Chamaephytes is a feature of the flora 

 that agrees with the few other biological spectra that have been published for 

 Southern Hemisphere floras. The Chamaephyte protects its buds by the shelter 

 of the plant itself or its dead remains heaped around them. These persistent 

 aerial portions also serve as a break that arrests movement of unstable soil. 

 In the case of Hemicryptophytes protection is gained as a result of the sub- 

 terranean position of the renewal buds which lie in the upper crust of the soil. 

 This method is less suitable for unstable soil, which may "drift," exposing the 

 buds. Also, in dry weather the dead remains are likely to be blown away 

 instead of remaining heaped over the buried portions. It must be remembered 

 that in South Australia the adverse period is a time of drought and heat, not cold. 



Hemicryptophytes (H) are represented on the Pearson Islands by a single 

 species, the fern Cheilanthes tenitifolia (H, 1-9 per cent.). None were found 

 on Franklin Island and only 4 per cent, at Ooldea. This rarity of Hemicrypto- 

 phytes in the floras of the three arid districts of Australia, which have up to 

 the present been investigated on Raunkaier's system, is a feature requiring 

 further examination. 



No geophytes were observed. This may be due to the season of the 

 year, but it may be recalled that only a single species was observed at Ooldea 

 in the winter months. 



The therophyte flora (Th, 282 per cent.) is less than that at Ooldea 

 or Franklin Islands, but the percentage is still double that of the normal 

 spectrum. It is probable that the nvmiber of annual species would be increased 

 were it possible to visit the islands during the wet season, though comparisons 

 with Franklin Islands are quite justifiable since those islands were also visited 

 in January. Probably the percentage of therophytes on Pearson Islands is 

 less than that of the other two districts because, owing to the topography of 

 the islands, denudation is more rapid and the soil correspondingly less stable. 



Appendix I. 

 ANALYSES OF SOIL SAMPLES FROM PEARSON ISLANDS. 



By J. G. Wood, B.Sc. 



The samples were all obtained from a section through the first 4 to 

 6 inches of the soil after brushing away the surface layer. They 

 were, all but No. 1, derived from granite, and are of the "coarse sand" 

 type, i.e., the majority of the particles have a diameter of from 1 to 02 mm., 

 and consequently the soils have little water-holding capacity. This is modified 

 in the case of soil No. 6, which has a considerable amount of clay, and also in 

 Nos. 2 and 3, which have 99 per cent, and 139 per cent, of humus, respectively. 



Before analysis the soils were air-dried and then passed through a 10-mesh 

 (2-6 mm.) sieve, and the sieved soil was finely ground and used for analysis. 

 The screenings consisted of fragments of granite, felspar, quartz, and in some 

 cases (mentioned below) dried vegetation as twigs, roots, etc., and also drop- 

 pings from cockroaches. The screenings in each case amounted to about one- 

 third of the total sample of soil. 



The results of the analyses are set forth in the table : — 



(11) Bews, J. W., The growth forms of Natal Plants, Trans. Rov. Soc. S. Africa, v., 

 p. 624, 1916. 



