JUNCACEAE 477 



Distichia, Oxychloe, Patosia, by dicliny or dioecism). Homogamy seldom occurs. 

 Schulz ('Beitrage/ II, p. 171) describes Juncus squarrosus L. as homogamous *as 

 the anthers of the three stamens dehisce immediately after the opening of the flower.' 

 Kirchner ('Neue Beob./ p. 10), however, says that the species is feebly protogynous. 

 He states that Juncus arcticus Willd. is decidedly homogamous (at Zermatt), but with 

 such short filaments that the tips of the anthers do not reach the stigma, self- 

 pollination being thus prevented. 



In protog}'nous species, the length of the first (female) stage of anthesis varies 

 greatly; according to Buchenau (' Bestaubungsverh.,' p. 367) from a few hours 

 (Juncus tenuis Willd.) up to several days (species of Luzula flowering in spring). 

 After this female stage there is generally a hermaphrodite one, though in some 

 species (Luzula campestris DC, L. spadicea DC), the stigmas have long been 

 withered when the anthers dehisce, so that first a neuter and then a male stage 

 follow the female one. A period sometimes follows closely on the hermaphrodite 

 stage during which the stigmas are still receptive when the pollen has all been 

 scattered and in chasmogamous flowers the perianth has closed. Then, if a part 

 of the stigma is shut in, automatic self-pollination may take place by means of the 

 pollen remaining on the perianth leaves. 



Cleistogamy sometimes occurs. F. von Miiller states that the Australian species 

 Juncus homolocaulis F. MuelL, and perhaps also the North American species 

 J. setaceus Rostk. and J. repens Mtchx., seem to flower entirely cleistogamously. 

 Cleistogamous flowers may also be found in Juncus bufonius Z. (frequently), 

 J. Chamissonis Knuth, J. capitatus Weigel, J. capillaceus Lam., Luzula purpurea 

 Link (occasionally), and probably in other species. 



In cleistogamous flowers, the style and stigmas must be but short. As a rule 

 the three inner stamens also disappear Buchenau ('Bestaubungsverh.,' p. 371) states 

 that the importance of their absence lies in the fact that the opening of the perianth 

 depends upon the turgescence of the floral receptacle, including the bases of the 

 stamens, so that when the inner ones are absent, this turgescence is lessened. 



An extremely curious thing in many species is the intermittent flowering. 

 Buchenau (* Monographia,' p. 41) states that during anthesis a large number 

 (one-fifth up to even one-third) of the flowers are found open on one day, then, 

 in spite of favourable weather, not a single open one for 10, 12, 14, or even 

 21 days, then again a large number open on one day, and so on. No connexion 

 between this mode of flowering and the weather can be traced. It does possess 

 oecological importance, however, for it greatly assists crossing by the numbers of 

 flowers open at the same time (Buchenau, ' Bestaubungsverh.,' p. 369). This may 

 also explain the fact that this method of flowering does not occur in species 

 with crowded inflorescences, such as most of those belonging to Luzula, while it 

 is seen most plainly in species of Juncus bearing single flowers, and in those with 

 few-flowered capitula. 



Buchenau (' Monographia,' p. 41) describes the course of anthesis in large 

 inflorescences, and also inside the single capitula, as ascending; but the terminal 

 spikelet or the terminal capitulum is mature before the lateral ones. Also where the 

 flowers are solitary the terminal one is mature before the associated lateral ones. 



The length of anthesis is (Buchenau, ' Bestaubungsverh.,' p. 370) very variable, 



