Mr. Hassall on the Structure of the Pollen Granule. 567 



a close inspection will disclose to be not identical in their 

 structure. 



The same type of granule is frequently found to extend 

 through more than one allied order, and sometimes through 

 a whole alliance ; thus, among monocotyledons, we find one 

 form extending, with slight variations of size, &c., not only 

 through the group Glumosce, but likewise through the Alli- 

 ance Typ hales and the genus Juncus, Another distinct form 

 is met with in the genera Butomus and Tradescantia, from 

 these passing through Liliales, looiales, Bromeales and iVar- 

 cissales ; and lastly, a third formation of granule occurs which 

 connects the orders of the Alliance Amomales. Among dico- 

 tyledons the three-lobed form prevails through the obviously 

 connected orders Rosacea, Pomea, Amygdalem and Saxifra- 

 gacecB. The same type, differing only in exact form and size, 

 unites the Alliances Scrophulales and Solanales. Papaveracece 

 and Cruciferce are also joined with each other in the form and 

 structure of their pollen granules. 



It is unnecessary to enumerate in this place the various 

 orders and sections of orders which are characterized by the 

 possession of a pollen granule of a peculiar and distinctive 

 form, as all the information which can at present be given upon 

 this subject may be learned by reference to the lists of plants 

 examined under each order already mentioned. Further re- 

 search will doubtless extend the number of these orders, clear 

 up many apparent anomalies, and also, it may be reasonably 

 expected, add somewhat to the number of exceptions. 



The truth of Mohl's second and third assertions, contained 

 in the statement in the preceding page, I would most di- 

 stinctly deny, more especially that of the last ; and in doing 

 so I would observe, that I rely solely on the evidence which 

 my own experience furnishes. Out of 366 genera submitted 

 to the microscope, two exceptions only have occurred to the 

 rule, that species of the same genus possess the same type of 

 pollen granule ; but I have little doubt future investigation 

 will explain these isolated exceptions. The first exception 

 occurs between two species of Linum, L. usitatissimum and 

 L. africanum ; in the former the granule is somewhat square, 

 and the extine contains six perforations for the escape of pol- 

 len tubes ; in the latter it is three-lobed, with three pollen 

 tubes : the second is found in the genus Viola. In Viola 

 tricolor the pollen is an elongated six-sided figure, emitting 

 four pollen tubes, and in Viola montana and V, odorata it is 

 small and three-lobed. 



If the above statements be correct, it would appear that an 

 examination of the pollen granule establishes and confirms in 

 a remarkable manner the naturalness of many orders and 



