A NATUMAlISrS WANDEBINOS 



its Itx-ked-up petals, I found the labellum beautifully marked 

 with lines of purple, carmine and orange, and the column also ; 

 hut no insect eye could eter be fascinated or allured by its 

 painted whorls. 



In the rather inconspicuous Goodj/era 2)rocem self-fertilisa- 

 tion takes place by the swelling up of the viscid matter of the 

 stigma beyond its true boundary, till it touches, as seen in 

 Fig, 28, the viscid disk of the poUinia, and spreads into tlie 

 poUinia chamber. I have no doubt this takes place in many 

 other species of Qoodyara^ and very probably also in our own 

 Highland si}ecies, Good^era repens. Other species which I have 



nr.. 27, no. 28. 



GOODYEEA PROCERA. ; A, BWOLLES tJP CArDICUEfi OF POLLFNIA f SOMEWHAT ESAd- 

 GEBA-rO)) ; B, 6PUT BOSTEJ^LUM, SHOWING IS nO. 28 THE DISK OP FOLLINIA ; 

 C, 6TI0HA : tl, UPPEa HABGIH OF STIGMA BEFORE BTlaUATiC FLUID UAH BEQCN 

 TO SWELL ; B, THE mQUATtO fttW SVfOLUOt UP, 



not been able to designate by name presented similar or allied 

 modifications for securing self- fertilisation. 



To me was especially interesting the purple Amndinaj 

 which one might imagine to have become tired of vainly 

 displaying its beauty to wayward and inappreciate butterflies 

 and bees, and had assumed a form that should — let all the 

 glittering humming wiugs pass heedless as they wouid^ — per- 

 petuate a fertile race. 



These instances go to show that the rule that " the flowers 

 of orchids are fertilised by the pollen of other flowers " is not 

 so universal as has been supposed. It is to be feared that too 

 often the interesting cases of flowers observed to be cross- 

 fertilised by insects have been recorded, while those of flowers 

 otherwise fertilised have not been mentioned, so that the law 



