Ihvu pollinated In the time the 

 (towels have opciK-il. rWs is in direct 

 couirast to the supposition by 

 Nkholls and it leaves Ihe way open 

 lor natUlttl hybridization lo OCCVJ in 



Thelympra v&ma, nw follow me. 



observations however will show thai 

 this can onlv Occur with 7. wnasa 

 participating as Hie female parent of 

 the cross. 



The absence ol pollen from the 

 anthers ol 20$| ol the open flowus 

 can be accounted for in two wax-, 

 a The species may occasion- 

 ally be c loss pollinated and 

 insects have removed the 

 pollen. 



b These dowers are male stei 

 ilc. 



\lv conelusutu is that the Mowers 

 ate male sterile and is based on the 

 following observations. 



I. The polhnarv mechanism is 

 completely gcaied lo sell-pollination 

 and is in no wav adapled for in- 

 sects because. 



a The pollima are fiiablc. 

 mealy and quickly lose any 

 coherence 

 h I he pollinia are free from 

 each other and there is no 

 visudium uniting them for 

 insect removal. Thus if there 

 was an insect agency n would 

 remove one or two of ihe 

 pollinia and leave the rest 

 My observations show thai 

 the anthers were completely 

 empty and the stigmas Were 

 free ol pollen. It would he 

 a F oor cross pollinating sys- 

 tem that rohbed the anthers 

 hut did not deposit pollen on 

 Ihe stigmas, 

 e The rostellum is poorly deve- 

 loped and does not effectively 

 shield the stigma from the 

 anther. 



2. rhe anther sacs of some im 

 qpcrlcd buds that I dissected were 

 CjUlW emply eliminating Ific possi- 

 bility ot insect removal. 



Ihcse observations prove con- 

 clusively thai Ttwlymttfa vtnosa k 

 predominately a self * pollinminj 

 species. However approximately lo- 

 ot lite population at lake Mountain 

 is male sterile and since the stiymav 

 of these are still functional they c;m 

 In Brass pollinaled bv other specie-, 

 I am sure that none are cross- 

 pollinated on I ake Mountain sinq 

 I base noi seen any oilier Thftunit*. 

 species in the area and certainly none 

 Powering in late l.tnnaiv MQWfiVg 

 whe e / nbHMa is crowing with a 

 normal! v cross pollinating speciev 

 e.g., as at Rocky Plain with tjlty 

 miiru Luuiile.t, then it ts feasible 

 thai a pollinating agent bearing pol- 

 linia of T. i.Mt'uhs could visit 7 

 tftRQSH especially as Ihcv grow intcr- 

 inm«lcd together, The chances aic 

 low that it will sinU- a llower with 

 nu pollen on the stigma, howevei 

 lilts uifrcquency will be offset bv the 

 large numher of ovules products' 

 f'om each cross. 



( o nments made by T. U. Muir 

 ..ud A t Beauglehole indicate In... 

 UK possihle hvhnd is not as common 

 as either parent T. vtnoas however, 

 is especially abundanl m the StH 

 and from the results of the fore- 

 going study. Ibis seems an csseniwl 

 prerequisite If il is to participate in 

 natural hybridization the frequent 

 of T iur,ul<s will noi be as ipipftf 

 ant but naturally any increase in if* 

 abundance will increaw the cfunw 

 of the cross occurring 



I hope to study this natural asso- 

 ciation lurlher at Rockv Plarn in the 

 co rang seasons. 



218 



V<1 Nat Vol gt 



