IN JAVA. 



95 



up and inundating, by the cliannels at thcs side of the rostellum, 

 at least the most external pollen masses on each side. These 

 poUinia emit their tubes over the rim of the rostellum, almost 

 obliterating it, into the stylary canal. On the opening of the 

 flower and the retraction of the anther-case, the most internal 

 poUinia may sometimes be found in the condition of loose 

 grains unaffected by the inundation of viscid matter. In its 

 fertilisation this species of Eria seems to resemble Dendrohium 

 clirysanihum. 



The mode of fertilisation described as occurring in Oplirys 

 apifera by Mr. Darwin, I found to be followed very closely by 

 a species of Eria near to E. javensis, in which the anther-cap 

 shrivels up backwards after rupturing, so as to disclose the 



PIG. 26, 



FIG. 26a. 



riG. 26.— CHKTSOGLOSSUM SP. ; A, ANTHER-CAP ; B, POLLISIA IS SITU ; 0, 



stigma; d, uppek mabgin of stigma; e, lowek margin op 



STIGMA. FIG. 26a. — SEOTIOK OP SAME. THE VISCID MATTES FLOWS 

 OVER THE MARGIN, D, INTO THE ANTHER. 



pollinia, which at once, even when quite shaded from wind and 

 all other disturbances, begin a slow tortuous movement, during 

 which they fall into their own stigmas, as seen in Fig. 23, p. 93. 

 In a species of terrestrial orchid unknown to me, but nearly 

 related, if not belonging to the genus Chrysoglossum, I 

 found these contrivances for effecting self-fertilisation carried 

 to their extreme limit, by its fertilising itself without ever 

 opening its florets at all (Figs. 25, 26). I observed them in 

 the forest, as well as grew a few of them in Mr. Lash's garden, 

 and every specimen was fertilised in the same way. In opening 



