32 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



its parents. I believe most of these hybrids are wild, and that 

 very few hybrid Saxifrages have been raised in cultivation. 

 Saxifrages may be classified, some as protandrous and some as 

 protogynous. I give a sketch of each of the two types, copied 

 from woodcuts in Hermann Muller's " Alpenblumen." But in 

 diplostemenous flowers the two rows of stamens do not open 

 and shed the pollen at the same time, and I incline to believe 

 that what insects generally do with Saxifrages is to convey the 

 pollen to the stigma of the same flower, and that cross-fertilisa- 

 tion is much less frequent here than in the heterostylous 

 isostemenous Primulas. This is a question to which further 

 observation may be suitably directed. But if I am to keep 

 within the limits of the time that has been allotted to me, I must 

 conclude. I have asked or suggested a great many questions. 

 Of course I do not expect them to be answered off-hand on the 

 spot by members of the present company ; but I think you will 

 agree with me that the genus furnishes a field for a wide range 

 of observation, deduction, and speculation 



A CLASSIFIED LIST OF CULTIVATED SAXIFEAGES, AEEANGED 

 AFTER ENGLEE'S MONOGEAPH. 



Section Bergenia , Moench (Megas^ea, Haw.) 

 (excluded from Saxifraga by Engler). 



crassifolia, Linn. Siberia. 



cordifolia, Haw. 



ligulata, Wall. Himalayas. 



ciliata, Wall. 



thysanodes, Lindl. 



Stracheyi, Hk. fil. & Thorns. Himalayas. 

 Milesii, Baker. 



purpurascens, Hk. fil. & Thorns. Himalayas. 



Section 1. — Cymbalaria, Griscb. 

 3. Cymbalaria, Linn. Asia Minor, Persia, W. Himalayas. 



Section 2. — Tridactylites, Haw. 



7. tridactylites, Linn. N. temp. zone. 



8. adscendens, Linn. Europe, N. America. 



