204 MR. Ch BBXTHA.M 0> T EUPHOBBIACEA 



2. Stenolobe^e. 



This small tribe was established by Mueller in the ' Prodromus,' 

 but absolutely rejected by Baillon, chiefly because it rests on a 

 character which can rarely be observed, and is therefore perhaps 

 not constant, and which is combined both with uniovulate and 

 with biovulate cells of the ovary, a difference considered of 

 primary importance in separating tribes. This main character 

 of Stenolobese, the linear embryo with narrow cotyledons, is an 

 essential, not an adaptive one likety to be affected by external 

 influences of soil, climate, or social conditions ; and it is strictly 

 geographical (southern, extratropical, and almost exclusively 

 Australian), thus giving strong evidence of the natural affinity 

 (consanguinity) of the genera in which it is developed. Its 

 supposed constancy has, moreover, been more and more con- 

 firmed as observation has extended. I have myself examined 

 the seeds of the great majority of the genera of Euphorbiaceae, 

 and we know the structure of others from reliable observations 

 of other botanists ; and although there may be in other tribes a 

 few embryos exceptionally small in comparison with the albumen, 

 or with exceptionally large and fleshy cotyledons, yet there are 

 none assuming the peculiar form of the Stenolobese. Even in the 

 nearest approach to it, in two or three species of the South- 

 African Adenocline and Seidelia, the cotyledons are still flat and 

 nearly twice as broad as the radicle, whilst in Stenolobese I have 

 never found them half as broad a^ain as the radicle. If in Buxus 

 and Daphniphjllum the cotyledons are narrow, still the general 

 form of the embryo is very different from that of the Stenolobese. 

 We have therefore maintained Mueller's primary series Steno- 

 lobese as one of our principal tribes, as detailed in my ' Flora 

 Australiensis,' to which I would refer for further particulars 

 as to the genera, We have, however, added to the tribe one plant 

 which, though not Australian, belongs to that South- Andine 

 region which, in many other instances, shows a connexion with 

 the Australian flora. This is the Dysopsis, Baill. {Molina, C. 

 Gay), from Chili and Juan Fernandez, which has so little appa- 

 rent affinity with the Crotonese, to which it had been technically 

 referred, that it has been conjecturally named as belonging to 

 very different orders, and even figured as a Hydrocotyle. It has, 

 however, precisely the embryo, as well as other characters, of 

 Stenolobese, though even there it constitutes a genus not closely 

 allied to any other one. 



• 



• 



