36 THE JOURNAL OF BOTANY 



with sticky pollen. The filaments are inclined outwards almost 

 parallel with the petals, so that the anthers stand well beyond the 

 utmost fringe of the feathery styles in all the plants I have yet 

 examined. Some photographs taken by a friend at South Perth 

 show the anthers within the periphery of the styles. The styles — 

 three split up dichotomously into innumerable branches — form a 

 downy-looking tuft upon the deep green globular ovary. The 

 blossoms unfold, several simultaneously, at intervals of two or 

 three days. The flower is without nectar. Though I have in the 

 aggregate spent many hours among the plants in bloom, I have 

 not yet seen a single flower visited by an insect. I believe, how- 

 ever, that the flower is entomophilous, for which belief I will give 

 reason later on. 



Flowering is over early in September, and fruits are ripe early 

 in October. Then the aerial part of the plant vanishes as if by 

 magic; what becomes of it I have not yet ascertained. In October 

 1910 I carefully examined altogether thirty-three plants in three 

 different localities about five miles apart at the angles of an equi- 

 lateral triangle. They had borne between them nine hundred 

 flowers, and produced eighty-three fruits (9-23 %). The highest 

 number of flowers borne by one plant was ninety, and the 

 lowest two. 



I will now briefly outline the annual life-history of D. stricti- 

 caulis. This species grows in clayey soil where the ground is 

 sopping wet during the growing season. In this it differs from 

 D. macrantha, which usually occurs on sand, but seems quite at 

 home in almost any soil, wet or dry. In the summer months this 

 same ground gets very dry — dust dry, in fact. So the plant is 

 exposed to great extremes of moisture conditions every year of its 

 life. Now, though the ground gets wet soon after the commence- 

 ment of the winter rains (about May), the plant remains beneath 

 the surface till early in August. It first appears as a thick 

 green tuft of leaves closely appressed with tentacles involute 

 (fig. 5). The stem elongates rapidly. It has attained full height 

 (about one foot) and commenced to bloom early in September 

 (fig. 4). Its subsequent history differs but little from that of 

 D. macrantha. 



The flowers, borne in corymbs of scorpioid cymes, are pink, 

 varying in shade, but usually bright rose. Individually the 

 flowering plant is, I think, more beautiful than D. macrantha, 

 but collectively I have not seen an effect produced by it even 

 approaching the splendour of D. macrantha plants in full bloom. 

 Its surroundings detract from rather than enhance its beauty ; it 

 usually grows under the shelter of "Jam Gum" trees {Acacia 

 acuminata Benth.) in places devoid, or nearly so, of under shrubs. 



The ephemeral pink blossom opens about 9 a.m. and closes at 

 sunset. It exhales a faint but sweet somewhat lemon-like odour, 

 which becomes stronger in still warm air, but remains pleasing 

 and unaltered in character. The floral structure is closely similar 

 to that of D. macrantha ; but the stamens are erect and the 

 anthers always well within the periphery of the styles ; usually 



