influence of the winter, it becomes entirely deciduous. 
Its foliage has a striking resemblance to the beautiful 
Pyrus japonica of Thunberg, and aflfords a strong argument 
in favour of cancelling the genus Choenomeles, which was 
formerly proposed in the Transactions of the Linnaean 
Society, chiefly upon Thunberg's assertion, that its fruit, 
when ripe, splits into 5 valves ; an erroneous statement, of 
which it is difficult to account for the origin. We have 
examined perfect and ripe fruit, produced in the neighbour- 
hood of London, with fully formed seeds, and we have 
foimd that it differs from Cydonia only in the seeds not 
having a mucilaginous testa. 
The fruit of this species is said to be large, but not 
applicable as food under any management. It is oblong, 
of a pale citron colour, with dry, woody flesh, and very 
austere juice. It has not, we believe, been ripened in 
this country. 
Brandies hairy, finally becoming smooth. Leaves 
stalked, stipulate, coriaceous, oval, acute at each end, 
finely serrated, half-folded together, above green, shining, 
generally tinged with red, beneath downy, with a downy 
stalk, which is channelled, and glandular in two rows 
above; stipules pale, with glandular teeth, the length of 
the petiole, 3-parted, with the middle lobe ovate erect, 
lateral spreading with auricles at the base on the outside. 
Flowers terminal, solitary. Ovary fusiform, smooth, often 
bearing ;f bractea above its middle. Calij.v 5-parted, re- 
flexed, downy inside. Segments subulate, with glandular 
serrated ends. Petals spreading, on short stalks, oblong, 
rose-coloured. Styles 5, connate at base. Stigmas capitate. 
J. L. 
