Mr* J. Lixdley's Observations on Pomacccc, 89 



But in an order so strictly natural as this is, greater difficult) 

 is always to be expected in Boding characters for genera, than 



in those of which our knowledge is more imperfect, and whose 

 series of individuals may therefore be considered less complete. 

 There also appear to be some important modifications of struc- 

 ture to which the attention of botanists has not hitherto been di- 

 rected ; and they promise to afford better distinctions than hai e 

 yet been employed. 



The form of leaves has usually been considered a mark by 

 which certain genera might be distinguished. Sir .James Smith 

 has however justly pointed out the genera] insufficiency of these 

 differences even in their most decided form. Thus Sotbm with 

 pinnated leaves differs in scarcely any other respect from Pyrut, 

 where they are simple. Nor can the Cratagi with angular leaves 

 be distinguished from such as have a regular outline. Yet, en- 

 tire and serrated leaves are almost certain indications of diffe- 

 rent genera; P/iotinia integrifvliu offering the only instance to 

 the contrary. And the fruit of this, which has not yet been 

 seen, may determine it to be a genus distinct from that to which 

 I have referred it. 



Bracteae are generally subulate, quickly withering and falling 

 off. Jn Mespilus they adhere to the tube of the calyx ; and in 

 Raphiokpis are persistent and leafy. 



Inflorescence can rarely be employed even as a secondary 

 character ; for in Pi/rus we have all the gradations from a nearly 

 simple to a very compound form. Nevertheless, the nearly ses- 

 sile flowers of Mespilus distinguish it from Eriobotrija and Cra- 

 taegus. The great terminal bunches of Photinia are very unlike 

 the lateral flowers of the last genus. The scaly racemes of 

 Raphiokpis and the naked axillary ones of Chamtemeles are pe- 

 culiar to themselves. 



The limb of the calyx is usually cup-shaped and persistent ; 

 vol. xiii. x but 



