THE ORCHIDS OF NEW ENGLAND. gy 



tracted base of the lip is grooved, or with incurved margins, 

 the trough leading as a sure guide to the narrow orifice of the 

 nectary. The two anther-cells are widely separated, but little 

 divergent ; their lower ends projecting strongly forward, bring 

 the naked discs just into line with the orifice of the nectary. 

 The pointed tip of a pencil brought to the orifice neatly 

 catches the sticky discs and brings away the pollen-masses ; 

 when the movement, which is effected within a quarter or a 

 third of a minute, converges them just enough to make them 

 hit the broad stigma (which lies rather high) upon the re-appli- 

 cation of the pencil. The ' drum-like pedicel ' (seen in H. 

 Hookeri), is present in this species also, but reduced to a mini- 

 mum ; the movement which takes place appears to result 

 wholly from its change of form, the portion towards the anther 

 contracting most, and to be one of depression solely." 



Habenaria tridentata, Barton's Three-toothed Orchis, which 

 has already been mentioned as coming close to the True 

 Orchises, is the Orchis tridentata of Muhlenberg and the 

 Gymnadenia tridentata of Lindley, and through July and 

 August at the North, and sometimes as late as August in 

 Eastern Massachusetts, presents its single leaf and its few 

 greenish-white flowers.* This Habenaria resembles H. hypcr- 

 borca in that the anther-cells open "before the flower bud," 

 as Gray says, " is fully grown, or at least four or five days 

 before the flower opens, and as the flowers at this time are 

 horizontal and somewhat reclining, the packets of pollen 

 which spontaneously detach themselves from the pollen-mass 

 may fall out. In every case of flowers opening naturally, the 

 anther-cells were found widely gaping, yet so far as we can see 

 the pollen-masses cannot of themselves fall upon or reach the 

 stigmatic surface." There are, however, in this Habenaria, 

 " three club-shaped projections or processes, which are nearly 



*I have drawn them a little too large, in my illustration (Fig. 20). 

 7 



