rather longest, reflexed, and marked at the base witli 
a dark crimson patch. Filaments 10, united at the 
base, 7 bearing anthers, 3 sterile ones short and curved 
inwards. Pollen orange-coloured. Germen and aristw 
hairy. Style red, hairy at the base and smooth above. 
Stigmas 5, reflexed. 
This plant and the one we have described under the 
name of P. canescens are nearly related, and we sus- 
pect Mr. Andrews's figure was begun from the one and 
finished, from the other, as it is not like either, but is 
an intermediate kind of thing, and we believe some of 
his other figures to be in the same predicament, parti- 
cularly his Geranium astragaljfolium, the leaves of 
which are correct enough, but the flowers are not like 
it or any other that we have seen ; it seems to be quite 
imaginary. 
Our drawing was taken from a plant in the superb 
collection of Robert H. Jenkinson, Esq. where we ob- 
served it in flower this summer in greater perfection 
than ever we had seen it before ; it is rather more ten- 
der than some of the common kinds, but will thrive 
well if placed in an airy situation and not overwatered; 
and from its very distinct habit and foliage it makes a 
pleasing variety. The best soil for it is a mixture of 
turfy loam and peat, and a little sand to keep the com- 
post open. Cuttings soon strike root if placed on a 
shelf in the greenhouse. 
