1114 JOURNAL OF THE EOYAL HORTICULTURAL SOCIETY. 



review of previous work, the greater part of the paper is occupied by the 

 details of the selected types. The figures are good and numerous, and in 

 themselves indicate the stages of development ; thus, a series of 21 

 figures is devoted to the Pear, 17 to Cerastes Juliana, 11 to Agrimony, 

 and 10 to Sjnrcea Filipcndula. A summary of the more important 

 results is as follows : (1) Form of ovule-rudiment, and development of 

 nucellus. In Pyrece the papilla is symmetrical for a time, and the two 

 integuments are distinct in origin. In the other groups the ovular 

 papilla becomes more and more asymmetrical, the nucellus papilla is 

 displaced to one side, and the two integuments arise from adjacent cells. 

 (2) The integuments of the ovule and seed. Contrary to the opinion of 

 Baillon that all indigenous Bosacece (except Pyrece and Amygdalece) have 

 only one integument to the ovule, the author finds that almost all show 

 a double integument. In Geum, Fragaria, Potentilla, and Alcheinilla, 

 while the origin of two integuments is discernible, the inner becomes 

 suppressed, and the ovule is monochlamydeous. As a rule, the inner 

 integument arises from four epidermal cells, the outer from one sub- 

 epidermal cell. Their further growth depends on that of the nucellus 

 papilla. In Pyrece both integuments are distinct from chalaza to micro- 

 pyle. Where the papilla soon becomes asymmetrical, the two integu- 

 ments become more or less fused and can only be distinguished near the 

 micropyle, or not at all. With the early curvature of the papilla, there 

 is a corresponding suppression of one or both integuments on the side 

 next the funicle. The development of the seed-coats is variable in the 

 groups, both as to differentiation and cell- wall changes. (3) Evolution 

 of the nucellus. This arises very constantly from several sub-epidermal 

 cells. After the first stages of growth, there are considerable variations 

 in the different groups. (4) Evolution of the embryo-sac and its contents. 

 This is described in each case, but the details of fertilisation are not given. 

 The growth of the embryo and its relation to the endosperm are discussed 

 at length. The development of the seed is slow in comparison with its 

 size, and seeds may be found apparently full-grown, yet with only a tiny 

 immature embryo. — TP. G. S. 



Ovules as a Basis of Classification. 



Ovules considered as Bases of Classification. By G. Senii 

 (Bot. Zeit. No. 20, pp. 305-310, Oct. 16, 1902).— A review of a mono- 

 graph by Van Tieghem (Ann. des Sci. Nat. Bot., ser. 8, vol. 14, p. 213) 

 on a natural system of classification based on the primary reproductive 

 processes. A general outline of the resulting nomenclature is given, and 

 although the reviewer doubts the general practical applicability of the 

 system, he inclines to the belief that the development and conformation 

 of the ovum, as embracing the organs most protected from external 

 influences, might well be worthy of more consideration as a systematic 

 basis than it has hitherto been held to be. — C. T. D. 



Ghowth of Unfertilised Seed. 



Parthenogenesis in Thalictrum purpurascens. By M. Koernicke 

 [Bot. Zeit. No. 22, p. 343, Nov. 16, 1902).— Note on contribution by 



