NARCISSUS POETICUS AND ITS ALLIES 27 
lobed corona. Haworth’s description was taken from a garden 
plant. WN. dianthos was reduced to a aes of N. biflorus Curtis 
by Herbert, who, however, adiits that the plant was unknown to 
him; and it does not seem to have been subsequently described. 
A wild two- flowered Italian specimen in Herb. Mus. Brit., as well 
as similar exsiccata at Kew, more nearly = the typical 
French N. poeticus, but present broader more imbricated 
perianth- “nogmentee These are perhaps Ps age natural twin- 
flowered sports. similar form was lately sold by Messrs. Barr, 
probably | ars by bulb-division from solitary individuals 
h is same abnormality. 
rus Haw. is Sacttier 2-3-flowered form of which I have 
rane 
Curtis’s N. biflorus, which is placed next after N. iriflorus, 
was well known in Britain in the time of Parkinson. From its 
uniformly abortive anthers and lack of ovules it seems an unmis- 
takable hybrid rather than a real species or subspecies of this 
group, as it has sometimes been treated. Its imperfect anthers 
are depicted in Curtis’s plate, and are stil visible in much older 
th 
exsiccata, such as those of the Sloane Herbarium aworth and 
Herbert seem to have ae ies its phd iret sag latter ryt 
t its barrenness was due to long cultivati But it may be 
that 
questioned whether other Nereiog still fertile, te is been 
grown equally as long, and the plant is much more probably, as 
Barr thought, an ancient cross of some form of N. poeticus and 
N. Tazetta L. which has become widely spread owing to its 
exceptional vigour. Other slightly differing forms, perhaps not 
always barren, have been observed in Southern France in spots 
where N. posters s and N. aes: grow together, and of these N. 
Nas Haw e on 
Haworth’s Ga ea ‘N. stellaris, is of particular interest, for 
not only is it more adequately diagnosed in the pereen 2 than 
some of those preceding it, but it is fully described and well figured 
in Sweet’s British Flower Garden, published two years ae pea 
is clearly most akin to N. radiiflorus, with which it has 
Secor of Parkinson, which it pay well be, and it will be seen, 
8 
The N. latifolius yr of Clusius (Hist. Rar. Plant. 1. c.) a 
seems from it ‘ca angula”’ to be identical with this plant 
rather than with se I Salisb., and if this be admitted, 
