PETALOIDE^E. 



373 



with the rostellum, and thus liberate the 

 pollen-masses. These latter adhere firmly to 

 the insect's back by means of a gland at the 

 end of the stalk, so that the pollen-mass is 

 conveyed to another flower. It must, how- 

 ever, be remarked that if the pollen-mass 

 retained the nearly vertical direction it had 

 on its exit from the anther, it would, when 

 introduced by the insect into another flower, 

 strike against the anther, and not against 

 the stiguia. In order to place the pollen 

 in such a position that it shall impinge on 

 the stigma, the caudicle or stalk of the 

 pollen-mass contracts so as to give the pol- 

 len-mass the requisite horizontal direction. 

 This movement can readily be seen by 

 thrusting the point of a pencil into a flower 

 against the rostellum, when the pollen- 

 masses will adhere to the pencil, and may 

 be withdrawn from the anther-case, and, 

 if watched, will be seen to bend down- 

 wards, in the manner just described, im- 

 mediately after their 'removal from the 

 anther (fig. 464). These movements will 

 be again alluded to in the physiological 

 portion of this work. 



Fig. 464. 



Fig. 463. 



Column of Vanilla from the front and 

 from the side, the parts of the 

 perianth cut away : o, the anther ; 

 r, the rostellum; the 2 points to 

 the stigma. 



Orchis pyramidalis ; a, pollen-mass just removed from the anther, vertical ; 

 b, pollen-masses divergent and horizontal. 



Two distinct forms of the perianth sometimes present themselves on 

 the same flower-spike, so that the same species has received two specific 

 titles, and even three distinct generic names : e. g., Monachanthus, 

 Myanthus, and Catasetum, now all included in the last named genus, 

 and Cycnoches ventricosum and JEyertonianum, now known to be forms of 

 one and the same species. This was considered a most anomalous circum- 

 stance till it was shown by Darwin that the different forms represented 

 different sexes, the male flowers being different from the female. 



The OrchidacesB are terrestrial in temperate climates, forming subter- 

 raneous tubers or tuberously enlarged fibrous roots, from which the 

 flowering-stern shoots up afresh every season. In warm and moist climates 

 they are very frequently epiphytic, hanging on the branches of trees, or 

 even attaching themselves to rocks and other foreign objects. These kinds 

 generally form some kind of stem-tuber, either from the lower interncdes 

 of the axis which has just flowered, or of a new axis, sometimes from the 



