592 Barnhart : Heteromorphism in Helianthemum 



individual capsules from the petaliferous flower than in those from 

 the apetalous flowers, but it is unusual for all the ovules to mature 

 in either case. 



Following is a table showing the number of stamens and 

 ovules in each kind of flower in each of the species under discus- 

 sion. The material studied, in addition to my own herbarium, is 

 in the herbaria of the New York Botanical Garden and Columbia 

 University. ' The table is the result of numerous careful dissec- 

 tions, but all the species are not equally well represented, so that 

 in some instances these figures might be slightly altered by a fur- 

 ther study of better material. 





Petaliferous Flowers. 



Apetalous Flowers. 





Stamens. 



Ovules. Stamens. 



Ovules. 



H. Canadense. 



3° 



30-60 



4 



6-20 



H. majus. 



3° 



30-60 



4 



6-20 



H. capita turn. 



12-20 



8 



3 



3 



H. Nashii. 



IS 



8-10 



5 



3-6 



H. a rent cola. 



25 



20 



5 



IO 



H. corymbosum. 



25-3° 



20 



3-6 



9 



H. Georgianum. 



12-16 



35 



8 



' 17 



An examination 



of the 



available material 



of the 



remaining 



rth American spe< 



:ies fine 



ludinsr the Mexica: 



n ones) 



shows that 



H. glomeratwn, H. argentenm and H. Chihuahuense are hetero- 

 morphous, while H. scopariiim, H. Greenei, H. Mendocinense ■, H. 

 nutans, H. patois, and H. Pringlei seem to form quite a natural 

 group, with homomorphous flowers. H. Carolinianum is also 

 homomorphous, but otherwise shows a much closer relationship 

 with the heteromorphous species than with the group just referred 

 to. In fact, it seems to me like a heteromorphous species which 

 has lost its apetalous form of flowers ; and it would not surprise 

 me, if its origin could be traced, to find that it had heteromorph- 

 ous ancestors. 



Tarrytown, N. Y., October, 1900. 



