On THE CLASSIFICATION OF ASCLEPIADS. 
That notable multiplication of genera which marks the 
recent history of the Asclepiadacese has proceeded as it were 
in violation of one of the most firmly established principles 
of taxonomy ; the principle that plants are related in the 
ratio of the similarity of their fruit-structure; that plants 
whose fruits are essentially at one are of one genus. Now 
the peculiarly formed follicular pericarp, with its spindle- 
shaped mass of flat obovate imbricated silky-appendaged 
seeds, is so exactly one thing throughout all the 1600 or 
1800 species of this family that it plays almost no part at all 
in the delimiting of the genera. Bentham and Hooker, in 
their elaborate key to the seven tribes and one hundred and 
forty-six genera of the order, do not once mention the fruit. 
It is essentially the same thing in all. And if the synthetic 
value accorded to pods and seeds in other families were ad- 
mitted here, instead of nearly two hundred genera of Ascle- 
Piads we should have scarcely more'than the two or three 
that were recognized by Tournefort and by Linneus. 
It was only within the present century that men began to 
see that, in the treatment of these plants, an exception must 
be made as to the synthetic value of similarity in fruit. 
Even as late as about the year 1815, everybody, even the 
most accomplished systematists, presented the genus Ascle- 
pias as including not only our types of Acerates, Asclepiodora, 
oo Podostigma and Anantheriz, but also even Vincetoxicum, Sar- 
~ stemma, Oxypetalum, Gonolobus, Enslenia and Hoya, not to 
mention the representative species of as many other genera, 
as the genera of Asclepiads are now understood and accepted. 
This complete dismemberment in recent times of the 
Asclepias of the celebrated eighteenth-century and early 
See | ABT ROR PERITI 
Prrtonta, Vol. HI. Pages 231-256. Dec. 9, 1897. 
31 
