IN ORCHIDEi^ AND ASCLEPIADE/E. 529 



SO to speak, of its mass, which in some cases produced its 

 tubes lucrcl}^ on continued immersion in water. I even 

 found that in this species, in the f:;radual decay of the n-^'j 

 flower, where the parts remain soft, the rupture and pro- 

 trusion of tubes took place while the mass w\as still in its 

 original position, innnersed in the cell of its anthera/ The 

 tubes produced in this situation often acquire a great 

 length, but coming, immediately on their protrusion from 

 the mass, in contact with the membrane of the anthera, 

 their course is necessarily altered ; and in their new direc- 

 tion, which is generally upwards, they not unfrequently 

 arrive at the top of the cell, or even extend beyoiul it. 



Tn addition to the several species of Asclepias already 

 referred to, Cjinanchim {Viiicetoxicuiu) 7ii(jrum is the only 

 plant of this family in which I have observed the whole of 

 the appearances ; namely, the rupture of the mass, the 

 production and protrusion of the pollen tubes, their union 

 into a cord, with the course and entrance of this cord into 

 the cavity of the style. 



The present essay, therefore, as far as regards this 

 family, might with greater propriety have been entitled, 

 "On the mode of impregnation in the genus Asclepias." 

 It seems, however, allowable to conclude, that in all the 

 genera having pendulous pollen masses, the same economy, 

 slightly modified perhaps in some cases, is likely to be 

 found. But among those with erect pollen masses, there 

 are several in which more considerable diftcrences may be 

 expected. Of this section of the family I have hitherto 

 had the opportunity of submitting only one plant to careful 

 examination, namely, ITojjd carnosa ; and even here my 

 observations are incomplete. 



Tn Ilof/a carnosa I have never found the pollen tubes 

 produced, or masses ruptured, while remaining in their 

 original ])Osition ; but I have succeeded in producing these 

 effects by bringing them in contact with certain parts of 

 the corona. 



The rupture and protrusion of ])ollcn tul)es, Ihen, take 



' Tab. 35, fig. 11. 



34 



